JP3097753B2 - Human C3b / C4b receptor (CR1) - Google Patents

Abstract

The present invention relates to the C3b/C4b receptor (CR1) gene and its encoded protein. The invention also relates to CR1 nucleic acid sequences and fragments thereof comprising 70 nucleotides and their encoded peptides or proteins comprising 24 amino acids. The invention further provides for the expression of the CR1 protein and fragments thereof. The genes and proteins of the invention have uses in diagnosis and therapy of disorders involving complement activity, and various immune system or inflammatory disorders. In specific embodiments of the present invention detailed in the examples sections infra, the cloning, nucleotide sequence, and deduced amino acid sequence of a full-length CR1 cDNA and fragments thereof are described. The expression of the CR1 protein and fragments thereof is also described. Also described is the expression of a secreted CR1 molecule lacking a transmembrane region. The secreted CR1 molecule is shown to be useful in reducing damage caused by inflammation and in reducing myocardial infarct size and preventing reperfusion injury.

Description

DETAILED DESCRIPTION OF THE INVENTION In accordance with the provisions of United States Code 35, §202 (C), the United States Government hereby acknowledges that it has rights in this invention. Portions of the fund for this invention have been funded by the NIH.

1. Introduction The present invention relates to the C3b / C4b receptor (CR1) gene and its encoded protein. The present invention also relates to CR1
Nucleic acid sequence and fragments comprising 70 nucleotides thereof and peptides encoded therefrom or 24
It also relates to proteins comprising amino acids. The invention also provides for expression of the CR1 protein and fragments thereof. CR1 nucleic acids and proteins have use in the diagnosis or treatment of disorders involving complement function and various inflammatory and immune disorders.

2. Background of the Invention 2.1. Complement tissue Complement tissue is about 10% of globulin in human normal serum.
Is a group of proteins (Hood, LE,
Et al., 1984, Immunology, 2d Ed., The Benjamin / Cummings P
ublishing Co., Menlo Park, California, p. 339). Complement (C) plays an important role in mediating immune and allergic reactions (Rapp, HJ and Borsos, T, 1970, Molec
ular Basis of Complement Action, Appleton-Century
-Crofts (Meredith), Ner York). Activation of complement components leads to the generation of a group of factors that include chemotactic peptides that mediate inflammation associated with complement dependent disease. Sequential activation of the complement cascade occurs via the major pathway involving the antigen-antibody complex or by the alternative pathway involving recognition of specific cell wall polysaccharides. Functions mediated by activated complement proteins include lysis of target cells, chemotaxis, opsonization, stimulation of vascular and other smooth muscle cells and degranulation of mast cells, increased permeability of small blood vessels , Control of leukocyte migration, and functional abnormalities such as activation of B lymphocytes and macrophages (Eisen, H. et al.
N., 1974, Immunology, Harper & Row Publishers, Inc.Ha
gerstown, Maryland, p.512).

During the stages of the proteolytic cascade, a biologically active peptide fragment, anaphylatoxin C3
a, C4a, and C5a (WHO Scientific Group, 1977, WHO T
ech.Rep.Ser.606: 5 and references cited therein) are released from the third (C3), fourth (C4), and fifth (C5) intrinsic complement components. (Hugli, T.
E., 1981, CRC Crit. Rev. Immunol. 1: 321; Bult, H. and He
rman, AG, 1983, Agents Actions 13: 405).

2.2. C3b / C4b complement receptor (CR1) The human C3b / C4b receptor, termed CR1, is an erythrocyte,
Present on monocytes / macrophages, granule cells, B cells, some T cells, spleen follicular dendritic cells and glomerular podocytes (Fearon, DT, 1980, J. Exp. Med. 152: 20, Wilson, J.
G., et al., 1983, J. Immunol. 131: 684; Reynes, M., et al., 1985, I.
mmunol. 135: 2687; Gelfand, MC, et al., 1976, N. Engl. J. Me
d.295: 10; Kazatchkine, MD, et al., 1982, Clin.Immunol.Im
munopathol.27: 170). CR1 is a solubilized form of the receptor that specifically binds to C3b, C4b and iC3b is found in plasma having ligand binding function and the same molecular weight as membrane-associated CR1 (Yoon, SH and Fearon, DT, 1985, J. Im
munol.134: 3332). CR1 binds to C3b and C4b, which are covalently attached to immune complexes and other complement activators, and the consequences of these interactions depend on the cell type carrying the receptor (Fearon, DT And Wong, WW, 198
3, Ann. Rev. Immunol. 1: 243). Erythrocyte CR1 binds to immune complexes for transport to the liver (Cornacoff, JB, et al., 1).
983, J. Clin. Invest. 71: 236; Medof, ME, et al., 1982, J. Ex.
p.Med. 156: 1739). Is CR1 of neutrophils and monocytes uptake adsorptive intracellularly through surface-covered depressions (Fear
on, DT, et al., 1981, J. Exp.Med. 153: 1615; Abrahamson, D.
R. and Fearon, DT, 1983, Lab. Invest. 48: 162) or after-feeding of receptor activation by phorbol esters, chemotactic peptides, or proteins present in the extracellular matrix such as fibronectin and laminin. By action (Newman, SL, et al., 1980, J. Immunol. 125: 2
236; Wright, SD and Silverstein, SC, 1982, J. Exp.M
ed. 156: 1149; Wright, SD, et al., 1983, J. Exp.Med. 158: 133.
8) The binding complex is internalized. CR1 phosphorylation has a role in acquiring phagocytic functions (Changelian, PS and Fea
ron, DT, 1986, J. Exp. Med. 163: 101). These cells are
Although treatment with antibodies to R1 enhances the response of cells to near-optimal doses of pork weed genogen, the function of CR1 on B lymphocytes is poorly defined (Daha, MR, et al., 1983). , Immun
obiol.164: 227 (Abstr.)). CR1 on follicular dendritic cells is useful for the role of antigen presentation (Klaus, GGB, et al., 1980,
Immunol. Rev. 53: 3).

CR1 also blocks C3 / C5 conversion of the major and minor pathways and acts as a cofactor for factor I cleavage of C3b and C4b. This is CR1
Has a complement regulatory function in addition to its role as a receptor (Fearon, DT, 1979, Proc. Natl. Acad. Sc.
iUSA76: 5867; Iida, K. and Nussenzweig, V., 1981,
J. Exp. Med. 153: 1138). In the alternative pathway of complement activation, the bimolecular complexes C3b and Bb are C3 active enzymes (convertes). CR1 (and at high concentrations, Factor H) binds to C3b and also promotes the dissociation of C3b, Bb. C3b, CR1
The formation of (and C3b, H) renders C3b amenable to irreversible proteolytic inactivation by Factor I, resulting in the formation of inactive C3b (iC3b). In the main pathway of complement activation, complex C4b, 2a is a C3 convertase. CR1 (and, at high concentrations, C4 binding protein, C4bp) binds to C4b and also promotes dissociation of C4b, 2a. Binding renders C4b susceptible to irreversible proteolytic inactivation by Factor I and is cleaved into C4c and C4d (inactive complement proteins).

CR1 is a glycoprotein composed of a single polypeptide chain. Four allotype forms of CR1 are known, with a difference of increasing ~ 40,000-50,000 dalton molecular weight. The two most common forms are the F and S allotypes, also called A and B allotypes, having molecular weights of 250,000 and 290,000 daltons, respectively (Dykman, TR,
Natl. Acad. Sci. USA 80: 1698; Wong, W. et al., 1983.
W., et al., 1983, J. Clin. Invest. 72: 685) Two rare forms have molecular weights of 210,000 and> 290,000 daltons (Dy
kman, TR, et al., 1984, J. Exp.Med. 159: 691; Dykman, TR,
Et al., 1985, J. Immunol. 134: 1787). These differences clearly indicate a variant in the polypeptide chain of CR1, rather than a state of glycosylation. Because they cannot be eliminated by treating the purified receptor protein with endoglycosidase F (Wong, WW, et al., 1983, J. Clin. Invest. 72: 685).
They have been observed to biosynthesize the receptor allotype in the presence of tunicamycin (Lublin, DM, et al., 1986, J. Biol. Chem. 26
1: 5736). All four CR1 allotypes have a C3b binding function (Dykman, TR, et al., 1983, Proc. Natl. Acad. Sci.
USA 80: 1698; Wong, WW, et al., 1983, J. Clin. Unveat. 72:
685; Dykman, TR, et al., 1984, J. Exp.Med. 159: 691; Dykman
TR, et al., 1985, J. Immunol. 134: 1787).

Two non-overlapping restriction fragments of the CR1 cDNA were shown to cross-hybridize under high stringency conditions (Wo
ng, WW, et al., 1985, Proc. Natl. Acad. Sci. USA 82: 771.
1). Both cDNA probes also hybridize to a number of restriction fragments of genomic DNA, and most fragments are common to both probes (author). The presence of the repeated coding sequence in CR1 was confirmed by sequence comparison (Klic
kstein, LB, et al., 1985, Complement 2:44 (Abstr.)).
Furthermore, the CR1 gene shows that, for some genomic restriction fragments, the genomic probe lacking the coding sequence has a sequence that recursed due to cross-hybridization (Wong, WW, et al. 198
6, J. Exp. 164: 1531). In addition, DNA from individuals with large S allotypes had other restriction fragments that hybridized to this genomic probe when compared to DNA from individuals with F allotypes. This indicates that genomic sequence duplication occurs in the context of the high molecular weight CR1 allele (author above).

CR1 shows homology with complement receptor type 2 (CR2) (Weis, JJ, et al., 1986, Proc. Natl. Acad. Sci.
USA 83: 5639-5643).

2.3. CR1 abnormalities in human disease C on erythrocytes in patients with systemic lupus erythematosus (SLE)
Reduction in R1 expression was observed in Japan (Miyakawa et al., 1981, Lancet 2:49
3-497; Minota et al., 1984, Arthr.Rheum. 27: 1329-133.
5), United States (Iida et al., 1982, J. Exp. Med. 155: 14).
27-1438; Wilson et al., 1982, N. Engl. Med. 307: 981-986)
And Europe (Walport et al., 1985, Clin. Exp.
l. 59: 547; Jouvin et al., 1986, Complement 3: 88-96; Holme
1986, Clin. Exp. Immunol. 63: 41-48).
Investigators have reported from several geographical areas. When viewed as a group, patients have an average number of receptors per cell that is 50-60% of that of normal people. In earlier reports, the number of CR1 on erythrocytes changes inversely to disease activity, i.e., the lowest number during the worst signs of SLE, and the higher number was observed during remission of the same patient. (Iida et al., 1982, JE
xp. Med. 155: 1427-1438). Also, the number of CR1 probably indicates that serum levels of immune complexes, serum levels of C3d, and uptake of complement-activating immune complexes, and are placed on erythrocytes as "harmless bystanders" Is known to be inversely related to the amount of erythrocyte-bound C3dg (Ross et al., 1985, J. Immunol. 135: 2005-2014; Holme et al., 1
986, Clin. Exp. Immunol. 63: 41-48; Walport et al., 1985, Cli.
n.Exp.Immunol.59: 547). One patient with SLE lacking CR1 on erythrocytes was found to have autoantibodies to CR1 (Wilson et al., 1985, J. Clin. Invest. 76:
182-190). Coincident with a decrease in anti-CR1 antibody titer and an improvement in the patient's clinical condition and a partial regression of receptor abnormalities. Anti-CR1 antibodies have been detected in two other SLE patients (Cook et al., 1986, Clin. Immunol. Immunopathol. 38: 135-1.
38). Recently, acquired loss of erythrocyte CR1 in conditions of active SLE and hemolytic anemia has been shown by observing the early loss of transfused erythrocyte receptors (Walport et al., 1987, Clin. Exp. .Immunol.69: 501
-507).

Also, the relative loss of CR1 from erythrocytes has been shown in patients with human immunodeficiency virus (HIV) infection (Tausk, FA,
1986, J. Clin. Invest. 78: 977-982) and are also observed in patients with lepromatous swelling (Tausk, FA, et al., 1985,
J. Invest. Dermat. 85: 58s-61s).

Abnormal complement receptor expression in SLE is not limited to erythrocyte CR1. It is seen that a relative deficiency of whole cell CR1 of neutrophils and plasma membrane CR1 of B lymphocytes of SLE patients occurs (Wilson et al., 1986, Arthr. Rheum. 29: 739-747).

Some patients with SLE type IV nephritis have detectable CR1
All of the antigen is lost from podocytes, while SLE
Modest types of nephritis and membrane-proliferative glomerulonephritis I
In non-SLE species of proliferative nephritis, including type II and type II, CR1 expression on glomerular podocytes is not different from normal (K
azatchkine et al., 1982, J. Clin. Invest. 69: 900-912; Emanci
ipator et al., 1983, Clin.Immunol.Immunopathol. 27: 170-1.
75). However, patients with SLE type IV nephritis have a lower number of erythrocyte CR1 than SLE patients with nephritis or other types without nephritis (Jouvin et al., 1986, Co.
mplement 3: 88-96).

In vivo complement activation is dependent on neutrophil plasma membrane CR
1 Upregulate expression (Lee, J., et al., 1984, Cli
n.Exp.Immunol. 56: 205-214; More, FD, Jr., et al., 1986,
N. Engl. J. Med. 314: 948-953).

Complement activation is also associated with disease states with inflammation. Crohn's disease, an intestinal inflammation, is characterized by lymphoid infiltration of mononuclear and polymorphonuclear leukocytes.
It has recently been found that complement C4 levels in jejunal fluid of Crohn's disease are increased compared to normal controls (Ahrenstedt et al.,
1990, New Engl. J. Med. 322: 1345-9) Another condition associated with the complement system in inflammation is heat injury (burns, frostbite)
(Gelfand et al., 1982, J. Clin. Invest. 70: 1170; Demling
Et al., 1989, Surgery 106: 52-9), hemodialysis (Deppisch).
Et al., 1990, Kidney Inst. 37: 696-706; Kojima et al., 1989, Ni.
ppon Jenzo Gakkai Shi 31: 91-7), post-pump syndrome of cardiopulmonary bypass (Chenoweth et al., 1981, Complement Infla.
mm. 3: 152-165; Chenoweth et al., 1986, Complement 3: 152-.
165; Salama et al., 1988, N. Engl. J. Med. 318: 408-14). Both complement and leukocytes have been reported to be involved in the pathogenesis of adult respiratory distress syndrome (Zilow
Et al., 1990, Clin. Exp. Immunol. 79: 151-57; Langlois et al., 1
989, Heart Lung 18: 71-84). Activation of the complement system has been suggested to be involved in the development of fatal complications in sepsis (Hack et al., 1989, Am. J. Med. 86: 20-26). Cochrane, 1984, Springer Seminar Imm
unopathol. 7: 263), glomerulonephritis (Couser et al., 1985, Kidn)
ey Inst. 29: 879), hemolytic anemia (Schreiber & Frank, 1)
972, J. Clin. Invest. 51: 575), myasthenia gravis (Lennon
Et al., 1978, J. Exp. Med. 147: 973; Biesecker & Gomez, 198.
9, J. Immunol. 142: 2654), type 2 collagen-induced arthritis (Watson & Townes, 1985, J. Exp. Med. 162: 1878) and empiric allergic neuritis (Feasby et al., 1987, Brain Res. 4).
19:97), resulting in tissue injury in animal models of autoimmune disease. The complement system has also been implicated in hyperacute allograft and xenograft rejection (Knechtle et al.,
1985, J. Heart Transplant 4 (5): 541; Guttman, 1974, T
ransplantation 17: 383; Adachi et al., 1987, Trans.Proc. 19
(1): 1145). Complement activation during immunotherapy with recombinant IL-2 appears to result in the severe toxicity and side effects observed from IL-2 treatment (Thijs et al., 1990, J. Immunol. 144:
2419).

Complement also plays a role in diseases involving the immune complex. The immune complex is rheumatoid arthritis or
Hematogenous malignancies such as SLE, AIDS (Tayler et al., 1983, Arth
ritis Rheum. 26: 736-44; Inada et al., 1986, AIDS Research.
2: 235-247) and autopathies and / or deficiencies associated with complement activation are found in a number of pathological conditions including, but not limited to (Ross et al., 1985, J. Immunol.
l.135: 2005-14). Inada et al. Show that erythrocyte CR1 has a functional role in removing circulating immune complexes in autoimmune patients, thereby inhibiting the deposition of immune complexes in body tissues (Inada et al., 1989, Ann. Rheum.Dis 4: 287). Decreased CR1 activity has been associated with clinical disease status in ARC and AIDS patients (Inada et al., 1986, AIDS Res. 2: 235).

3. SUMMARY OF THE INVENTION The present invention relates to the C3b / C4b receptor (CR1) gene and its encoded protein. The present invention also relates to CR1
A nucleic acid sequence and a fragment comprising 70 nucleotides thereof and a peptide encoded therefrom or
It also relates to a protein comprising 24 amino acids. The invention further provides for expression of the CR1 protein and fragments thereof. The genes and proteins of the present invention have use in the diagnosis or treatment of disorders involving complement function and various immune systems and inflammatory disorders.

Detailed embodiments of the invention, as detailed in the Examples section below, describe the cloning, nucleotide sequence and deduced amino acid sequence of the full-length CR1 cDNA and fragments thereof. Expression of the CR1 protein and fragments thereof has also been described. C3b
And / or expression of a CR1 protein and a fragment thereof comprising a binding site for C4b and exhibiting a Factor I cofactor function.

Also described in the following examples are soluble CR1
The production and purification of a molecule, which is shown to be therapeutically useful in treating inflammatory responses and reducing the size of myocardial infarction and preventing reperfusion injury.

3.1. Definition Ad2MLP = Adenovirus 2 late promoter C = Complement C3 (ma) = Methylamine treated C3 C4bp = C4 binding protein CMV = Cytomegalovirus CR1 = Complement receptor type 1, C3b / C4b receptor CR2 = Complement Body receptor type 2 DCFDA = Dichlorofluorescein diacetate HPLC = High performance liquid chromatography iC3b = Inactivated C3b LHR = Long homologous repeat mAb = Monoclonal antibody PAGE = Polyacrylamide gel electrophoresis RPAR = Reverse passive Arthus repeat SCR = Short consensus Repetitive sCR1 = soluble CR1 molecule 4. Description of the drawing FIG. Nucleotide and amino acid sequence of the entire CR1 coding region. This sequence is the octamer Ec of clone λT109.1.
Begins with the first nucleotide following the oRI linker.
Nucleotide number 1531 in this sequence is the first nucleotide 5 'to nucleotide number 1 in the sequence shown in FIG. The strand corresponding to the mRNA is shown, the deduced amino acid sequence is shown below. The putative signal sequence encoded by nucleotides 28-147 is bracketed.

FIG. 5.5Kb restriction map of human CR1 cDNA. Black bars indicate cDNA, restriction sites are H, Hind III; B, BamHI; R, Ec
oR I; P, Pst I; A, Apa I; S, Sac I; G, Bgl II; K, Kpn I. The cDNA clone from which the sequence was derived is shown below the map. Arrows indicate the direction and extent of sequence analysis by the dideoxynucleotide chain termination method. cDNA
Clones were oriented based on restriction maps and the identity of overlapping sequences.

FIG. 5.5 Kb nucleotide sequence of human CR1 cDNA. m
The strand corresponding to the RNA is shown, and base number 1 (corresponding to base number 1532 in FIG. 1) is the first base after the EcoRI linker of the most 5 'clone.
The stop codon is underlined. 110bp in the box
The sequence is found between nucleotides 147 and 148 (arrow) and is believed to be part of the intervening sequence.

FIG. Dot matrix analysis of the 5.5 Kb nucleotide sequence of human CR1 cDNA. Dots are plotted when there is a match of at least 40 bp out of 90 bp. The black line bisecting the square diagonally indicates sequence identity with itself. Two additional parallel black lines at 1.35 and 2.7 Kb from the line of identity are two directly tandem long homologous repeats (LHR) of 1.35 Kb each. Two LHR
The six thinner dotted lines in between correspond to short consensus repeats of 0.20.2 Kb. This short consensus iteration (SC
R) exceeds 0.4 Kb for long homologous repeats.

FIG. Deduced amino acid sequence of human CR1. Each residue is represented by a one-letter code (Lehninger, AL1975, Biochemistry, 2n
d Ed., Worth Publishers, Inc., New York, p. 72). Residues in long homologous repeats have been aligned to show their homology. All residues in LHR-B are shown,
The LHR-C and LHR-D residues are shown only where they differ from the LHR-B residues. The hydropathic profile is the COOH of the protein to indicate the putative transmembrane region.
It is arranged below the end. Lines are placed above the four positively charged residue portions immediately after the hydrophobic sequence. Protein kinase C phosphorylation site in epidermal growth factor receptor
Six amino acid sequences with 67% homology are underlined. A schematic of the CR1 protein is shown above the sequence. (T
M) transmembrane domain, (Cyt) cytoplasmic domain, (3'UT) 3 '
Untranslated sequence.

FIG. (A) SCR alignment of CR1. Repeating unit are put 1-23 and number towards the COOH-terminal from the NH ₂ -terminus. Space is provided to maximize alignment. Boxes represent invariant residues and vertical arrows indicate amino acid conserved portions. A residue is considered conserved if the residue or conservative substitution is present in at least half of the SCR. Horizontal arrows indicate SCRs also sequenced from CR1 genomic clone 2-38 and encoded by one exon. (B) shows a restriction map, a sequencing method, and a partial sequence of genomic clone λ2-38. The restriction site was (B) BamH
I, (S) Sac I, (E) EcoR V, (K) Kpn I, (P) Pst I
It is. Horizontal arrows indicate exon-intron boundaries.

FIG. Alignment of SCR consensus sequences for proteins known to have this structure. Spaces have been introduced to maximize alignment. Residues are considered conserved in FIG. 5 except for proteins with one or two SCRs where at least half of other proteins are conserved when present. It is. The underlined portions of CR2 and C2b indicate that no sequence information has been published in this region for these proteins. Boxes indicate invariant half cystine. The number to the right of the sequence indicates the number of the SCR used to generate the consensus sequence. Protein abbreviations and references for sequence data used to determine consensus sequences are as follows: (CR1) complement receptor type
1, (H) Factor H (Kristensen, T., et al., 1986, J. Imm
unol. 136: 3407), (C4bp) C4 binding protein (Chung,
LP, et al., 1985, Biochem. J. 230: 133), (CR2) complement receptor type 2 (Weis, JJ, et al., 1986, Proc. Natl. Acad.
Sci. USA 83: 5639), (Ba) a proteolytic fragment of factor B (Morley, BJ and Cambell, RD, 19).
84, EMBO J. 3: 153), (C2b) C2 proteolytic fragment (Gagnon, J., 1984, Philos. Trans. R. Soc. Lond. Biol. Sc)
i.306: 301), the γ subunit of (Clr) C1 (Leytus, S .;
P., et al., 1986, Biochemistry 25: 4855), (XIIIb) b subunit of factor XIII (Ichinose, A., et al., 1986, B
iochemistry 25: 4633), (β2GP1) β2 glycoprotein I (Lozier, J., et al., 1984, Proc. Natl. Acad. Sci. USA8
1: 3640), (Hap) Haptoglobin (Kurosky, A., et al., 19)
80, Proc. Natl. Acad. Sci. USA 77: 3388), (IL-2-
R) Interleukin-2 receptor (Leonard, WJ,
Et al., 1985, Science 230: 633). An asterisk indicates that an incomplete sequence is available.

FIG. Schematic diagram of the proposed structure for the human CR1 sequence. The COOH-terminal cytoplasmic region is to the right of the lipid bilayer.
30SCRs are linearly arranged outside the cells of the cytoplasmic membrane.
Parentheses indicate LHR. The inset shows one SCR enlarged to show the triple loop structure.

FIG. Restriction map of the insert of plasmid pBSABCD encoding human CR1. Within the box delineating the region containing the coding sequence, there are nine fragments of eight cDNA clones ligated to form a CR1 construct. Brackets indicate LHR-A, -B, -C, and -D, respectively. The line below the box indicates the location of the newly isolated 5 'cDNA clone. Restriction sites are A, Apa I,
B, BamH I; G, Bgl II; H, Hind III; K, Kpn I; M, BspM II; P, P
st I; R, EcoR I; and S, Sac I.

FIG. 10. Deduced amino acid sequence of a 5 'cDNA clone encoding the seven SCRs of LHR-A, and this sequence and LHR-A
Alignment of B, -C, and -D with the corresponding SCR. The four cysteines conserved in each SCR are underlined. The residues LHR-B, -C and -D are shown only where they differ from those of LHR-A.

FIG. Restriction maps of expression plasmids piABCD and pMTABCD. Pm _MT and _PCMV indicate mouse metallothionein and cytomegalovirus immediate early promoter, respectively.

FIG. PiABCD (panels a and b) and
Phase contrast of COS cells transfected with CDM8 vector alone (panels C and D) and indirectly stained with YZ1 monoclonal anti-CR1 antibody and fluorescein-labeled goat anti-mouse F (ab ') ₂ (panels a and c) ) And immunofluorescence (panels b and d) micrographs.

FIG. Analysis of C3b- and C4b-binding by COS cells expressing recombinant CR1. COS cells transfected with piABCD (panels a and c) or CDM8 vector only (panels b and d) were incubated with EAC 4b (lim), 3b (panels a and b) or EAC 4b (panels c and d). The rosette formation was examined with a phase contrast microscope.

FIG. Analysis by SDS-PAGE of recombinant CR1 expressed by transfected COS cells. CDM8 vector only (lanes 1 and 4) and piABCD (lane 2)
COS cells transfected in (5) and (5), and erythrocytes (lanes 3 and 6) from individuals with F and S allotypes of CR1 were surface labeled with ¹²⁵ I, respectively. The detergent lysate of the cells was subsequently immunosorbed with Sepharose UPC10 (lanes 1-3) and Sepharose-YZ1 (lanes 4-6) and the eluate was analyzed by SDS-PAGE and autoradiography under non-reducing conditions. analyzed.

Fig.15. ¹²⁵ I-C3 (ma) cleavage by Factor I in the presence of immuno-immobilized recombinant CR1. ¹²⁵ I-C3
Duplicate samples of (ma) were incubated with factor I in the presence of factor H (lane 1), lysate of COS cells transfected with CDM8 vector alone, Sepharose-UPC10 (lane 2), piABCD-
Sepharose-UPC10 (lane 3) pre-incubated with lysate of transfected COS cells, CDM8-
Sepharose-YZ1 (lane 4) pre-incubated with lysate of transfected COS cells, and piA
Treated with 6μ (lane 5), 12μ (lane 6) and 25μ (lane 7) of Sepharose-YZl pre-incubated with lysates of BCD-transfected COS cells. A sample of ¹²⁵ I-labeled C3 (ma) was also preincubated with lysates of piABCD-transfected COS cells in the absence of Factor I, Sepharose-YZ125.
μ (lane 8). After reduction, ¹²⁵ I-C3 (ma)
Was analyzed by SDS-PAGE and autoradiography.

FIG. 16. cDNA construct encoding CR1 deletion mutant.
The positions of the four LHR-encoding cDNA segments indicate the restriction sites used to prepare the deletion mutants.
Brackets are shown above the piABCD construct. The remaining cDNA restriction fragments in each mutant are indicated by black lines. Restriction sites are A, Apa I; B, Bsm I; E, BstE II; and P, Pst I.

FIG. 17. Comparison of the recombinant deletion mutant of CR1 with wild-type F and S allotypes of CR1. Red blood cells (lanes 1 and 7) and CDM8 vector only (lanes 2 and 8)
piABCD (lanes 3 and 9), piBCD (lanes 4 and 1)
0), COS cells transfected with piCD (lanes 5 and 11) and piD (lanes 6 and 12) were surface-labeled with ¹²⁵ I, respectively. 1-
6), immunoprecipitation with Sepharose-YZ1 anti-CR1 monoclonal antibody (lanes 7-11) and rabbit anti-CR1 antibody and Sepharose-protein A (lane 12), respectively. The eluate was subjected to SDS-PAGE and autoradiography under reducing conditions.

FIG. Cleavage of ¹²⁵ I-C3 (ma) by factor I in the presence of COS cells expressing full length CR1 and deletion mutants. Duplicate samples of ¹²⁵ I-C3 (ma) were prepared in the absence (Lanes 1-6) or presence of Factor I (Lanes 7-12)
In addition, only the CDM8 vector (lanes 1 and 7) respectively,
COS cells transfected with piABCD (lanes 2 and 8), piAD (lanes 3 and 9), piBD (lanes 4 and 10), piCD (lanes 5 and 11), and piD (lanes 6 and 12) were incubated. ¹²⁵ I-C
The samples of 3 (ma) were also factor H and factor I (lane 13), and only factor I (lane 1).
4) and incubated. After reduction, ¹²⁵ I-C3 (ma)
Analyzed by DS-PAGE and autoradiography.

FIG. 19. Schematic representation of the SCR form containing each LHR of CR1 and the predicted sites that determine the specificity of the C3b and C4b receptors. These secondary binding specificities are shown in parentheses.

Fig. 20. Schematic diagram showing the DNA region remaining in the soluble CR1 DNA construct. The full length CR1 cDNA region is indicated by the box at the top of the figure.

Fig. 21. Schematic diagram showing the main elements in the pTCS series of expression vectors.

Fig. 22. Schematic diagram of the expression vector pTCSgpt. The polyadenylation site was derived from the mouse Ig kappa sequence (NBRF Nucleic Database Accession # κcms, bp1306
-1714); Ad2 MLP and the tertiary segment are derived from the Ad2 sequence (N
BRF Nucleic Database Accession
# Gdad2, bp 5790-6069); SV40 early promoter is SV4
0 genome (NBRF Nucleic Database Accession # GSV40W). The gpt gene, the ampicillin gene and the bacterial origin of replication are in the vector pSV2
obtained from gpt (ATCC Accession No. 37145).

Figure 23. 4-20% SDS-P of antibody affinity purified sCR1
AGE. Non-reduction (lanes 1,2,3) and reduction (lanes 4,5,
6) Condition. Lanes 1,3: molecular weight marker; lanes 3,5: cell culture supernatant starting material; lanes 4,6: sCR1 purified by antibody affinity chromatography.

Figure 24. Cation exchange HPLC elution profile. The eluted protein was monitored by absorbance at 280 nm (Y axis). Flow-through (0-100 min) and elution sCR1 (150-
The absorbance at 160 min) was off the scale. The X axis represents the elution time (minutes).

Figure 25. 4-20% gradient SDS-PAGE of cation and anion exchange HPLC purified sCR1. SDS-polyacrylamide gels were run under non-reducing conditions. Lane 1, part of bioreactor supernatant; lane 2, part of bioreactor supernatant dialyzed against cationic HPLC starting buffer; lane 3,
Part of sCR1 elution peak from cation exchange HPLC column; Lane 4, part of sCR1 peak from cation exchange HPLC column dialyzed against anion HPLC starting buffer; Lanes 5 and 6, two parts of sCR1 eluted from anion HPLC Part of different fractions.

Figure 26. C5a induction of oxygen burst in human neutrophils. Following the C5a-induced enzyme burst, DCFDA was oxidized and fluoresced brightly. The fluorescence intensity measured by flow cytometry is shown on the X-axis and the number of cells is shown on the Y-axis.
Panel a, cell profile and gate; Panel b, C5a
0 minutes after addition; Panel c, 1 minute later; Panel d, 2 minutes later; Panel e,
3 minutes later; Panel f, 4 minutes later; Panel g, 20 minutes later. This DCFDA assay shows C5a with high sensitivity.

FIG. 27. Activation of human complement in the presence of sCR1 shows reduced C5a activity in the DCFCDA assay.
Panel a, unstimulated cells; Panel b, control without sCR1 showing high fluorescence; Panel c, sCR1 showing 75% decrease in fluorescence intensity.
DCFDA assay in the presence of. The Y axis is cell number and the X axis is fluorescence intensity.

Figure 28. Inhibition of classical pathway C5a and C3a production in human serum by sCR1. A similar profile was observed with antibody affinity purified or HPLC purified sCR1.

Fig. 29. Inhibition of complement-mediated hemolysis by recombinant sCR1. A similar profile is used for antibody affinity purification or HPLC-purified sC
Observed at R1.

FIG. 30. Macroscopic morphology of RPAR in sCR1-treated (left) and untreated (right) rats. (A) Both rats were intravenously injected with ovalbumin followed by sCR1 (left rat) or PBS (right rat) plus intact anti-ovalbumin (left part); 1/2 diluted anti-ovalbumin (middle part) or rabbit An intradermal injection of a mixture of IgG (right part) was received. Injections were performed in duplicate and the same results were obtained in the upper and lower rows. sCR1
Rats given no significant change, whereas untreated rats produced complete signs of RPAR.
(B) Skin surface of skin biopsy from (a). Biopsies from untreated rats (right) showed clearly visible lesions, whereas biopsies from sCR1-treated rats (left) showed normal morphology.

FIG. 31, Light micrographs of sCR1 treated (a) and untreated (b) rat skin biopsies. (A) Perivascular accumulation of polymorphonuclear and mononuclear cells was observed, but neither extensive infiltration of neutrophils nor emigration of red blood cells was observed. (B) Extensive infiltration of polymorphonuclear cells and emigration of red blood cells were confirmed.

FIG. 32. Clearance of sCR1 injected from rat and monkey blood showing a dual phase of α and β clearance phases.

FIG. 33. Autoradiography of Southern plots in which the CR1 cDNA and intron probe hybridize to an EcoR V digest of DNA from an individual expressing the F or F ′ allotype. The location of the HindIII fragment of lambda DNA is shown in kilobases on the left. The location of the F 'specific fragment is indicated by a single arrow.

FIG. 34. EcoR V restriction map of F allele of CR1. The white boxes indicate the locations of the exons, and the dotted boxes indicate the sites of hybridization of the intron probe. Brackets on LHR-B and -C indicate two regions where deletions may be made. V indicates EcoR V site.

FIG. 35. cDNA inserts into various forms of recombinant rCR1. Restriction sites shown are A, ApaL I; B, BamHI; C, Sa
c I; H, Hind III; L, Bgl I; P, Pst I; R, EcoR I; and S, Sma
I. The top schematic shows the CR1 protein, where SCRs with identical sequences are filled with the same pattern.

FIG. 36 Coomassie blue stained SDS-PAGE of recombinant sCR1 purified by YZ-1-Sepharose absorption under non-reducing conditions. Each lane is pasecABBCD (lane 1), pas
Contains 10 μg of recombinant sCR1 purified from the culture supernatant of COS cells transfected with ecABCD (lane 2) or pasecACD (lane 3). The position of the Mr marker is indicated by KD on the right.

FIG. 37. Cofactor activity of recombinant sCR1. Cleavage of the α 'chain of C3b was measured in the presence of increasing amounts of recombinant sCR1 from COS cells transfected with pasecABBCD, pasecABCD, or pasecACD.

FIG. 38. Inhibition of erythrocyte uptake of ¹²⁵ I-C3b dimer by recombinant sCR1. Erythrocyte binding ligand is C3b dimer, C3b
Monomer and pasecABBCD, pasecABCD, or pasec
Recombinant sCR1 from COS cells transfected with ACD
Was measured in the presence of increasing concentrations of

FIG. 39. Recombinant s purified from COS cells transfected with various plasmids encoding CR1 variants
Inhibition of alternative pathway (A) and major pathway (B) C3 convertase by CR1.

FIG. 40. Recombinant s purified from COS cells transfected with various plasmids encoding CR1 variants
Inhibition of alternative pathway (A) and major pathway (B) C5 convertase by CR1.

5. DETAILED DESCRIPTION OF THE INVENTION The present invention relates to the C3b / C4b receptor (CR1) gene and the protein it encodes. The present invention also relates to CR1
It also relates to the nucleic acid sequence and fragments thereof containing 70 nucleotides and the peptides or proteins containing the 24 amino acids they encode. The invention further provides
It also provides for expression of the CR1 protein and fragments thereof. Such CR1 sequences and proteins are of value in the diagnosis and treatment of disorders of the inflammatory or immune system and disorders involving complement activity.

In a particular aspect, the invention relates to soluble CR1 molecules and their expression, purification and use. As used herein, the term "soluble CR1 molecule" will mean the portion of the CR1 protein that is not expressed as a membrane protein on the cell surface, as opposed to the native CR1 protein. As a specific example, a CR1 molecule substantially free of the transmembrane region
R1 molecule. In a preferred embodiment, soluble CR1 molecules are secreted by cells that express them.

Detailed embodiments of the present invention, detailed in the Examples section below, describe the cloning and full nucleotide and deduced amino acid sequences of the full length CR1 cDNA and fragments thereof, and the expression of the CR1 product they encode. The expression of CR1 and fragments thereof that have a binding site for C3b and / or C4b and that inhibit Factor I cofactor activity is also described. The invention further provides
This is also explained by the production and purification of a soluble, truncated CR1 molecule. In specific examples, such molecules are shown to be therapeutically useful in reducing inflammation, and reducing myocardial infarction magnitude and preventing reperfusion injury.

5.1. Isolation of CR1 gene The entire coding sequence of CR1 gene and its deduced amino acid sequence are shown in FIG. Any human cell may serve as a source of nucleic acids for molecular cloning of the CR1 gene. CR1
Gene isolation involves isolating these DNA sequences encoding proteins that exhibit CR1-related structures or properties, such as C3b or C4b or immune complex binding, phagocytic modification, immunostimulation or proliferation, and complement regulation. Need. DNA is
The cDN can be synthesized from cloned DNA (eg, a DNA “library”) by chemical synthesis using standard techniques known in the art.
A by cloning, or by cloning of genomic DNA or a fragment thereof purified from the desired human cells (eg, Maniatis et al., 1982, Molecular
Cloning, A Laboratory Manual, Cold Spring Harbor La
boratory, Cold SpringHarbor, New York; Glover, DM
(Eds.), 1985, DNA Cloning: A Practical Approach, MRL P
ress, Ltd., Oxford, UK, Vol. I.II.). CR1 gene
Cells that serve as nucleic acid sources for cDNA cloning include monocytes /
Including, but not limited to, macrophages, granule cells, B cells, T cells, spleen hair follicular dendritic cells and renal glomerular podocytes. Clones derived from genomic DNA may contain regulatory and intronic DNA regions in addition to the coding region. Clones derived from cDNA will contain only exon sequences. Whatever its source, the CR1 gene should be molecularly cloned into a suitable vector for gene propagation.

In the molecular cloning of genes from genomic DNA, DNA fragments are generated, some of which
Would encode a gene. DNA can be cut at specific sites using various restriction enzymes. Alternatively, the DNase can be used in the presence of manganese to fragment DNA, and the DNA can be physically cleaved, for example, by sonication. The linear DNA fragments can then be separated according to molecular weight by standard techniques, including but not limited to agarose and polyacrylamide gel electrophoresis and column chromatography.

Once a DNA fragment has been generated, the specific DNA fragment containing the CR1 gene can be identified in a number of ways. For example, if a certain amount of the CR1 gene or its specific RNA, or a fragment thereof is available and it can be purified and labeled,
DNA fragments can be selected by nucleic acid hybridization with a labeled probe (Benton, W. et al.
And Davis, R., 1977, Science 196: 180; Grunstein, M. and Hogness, D. 1975, Proc. Natl. Acad. Sci. USA 72: 396.
1). Those DNA fragments that are substantially homologous to the probe will hybridize. If a purified CR1-specific probe is not available, the first selection method is
The CR1-rich nucleic acid fraction can be used as a probe. As an example, a probe that is a B cell cDNA with the message expressed by fibroblasts removed can be used. It is also possible to identify suitable fragments by restriction enzyme digestion and comparison of the fragment size with that expected according to known restriction maps, if available. After the initial selection, further selections can be made based on the properties of the gene or the physical, chemical or immunological properties of the expressed product, as described below.

The CR1 gene can also be identified by selection of mRNA by nucleic acid hybridization followed by in vitro translation. In this method, fragments are used to isolate complementary mRNA by hybridization. Such DNA fragments can be obtained using purified C
It can be R1 DNA, or DNA rich in the CR1 sequence.
Immunoprecipitation analysis or functional assays (e.g., for C3b or C4b binding, or promoting phagocytosis or immune stimulation, or complement regulation, etc.) of the isolated in vitro translation product of the mRNA identify the mRNA and thus identify the CR1 sequence. The containing complementary DNA fragment is identified. Furthermore, specific mRNA can be selected by adsorbing polysomes isolated from cells to CR1-specific immobilized antibodies.
Using the selected mRNA (from adsorbed polysomes) as a template, radiolabeled CR1 cDNA can be synthesized. The CR1 DNA fragment can then be identified among other genomic DNA fragments using radiolabeled mRNA or cDNA as a probe.

Methods other than isolating CR1 genomic DNA include chemically synthesizing the gene sequence itself from a known sequence, or CR1
This includes, but is not limited to, making a cDNA for the mRNA encoding the gene. For example,
As described above, RNA for the cDNA cloning of the CR1 gene was obtained using monocytes / macrophages, granule cells, B cells,
It can be isolated from cells including, but not limited to, T cells, dendritic cells, and podocytes. In a preferred embodiment, the tonsillar cells are cDNA cloned mRNAs.
Other methods are also possible, which serve as a source for (see Section 6.1.2, below) and are within the scope of the present invention.

The identified and isolated gene can then be inserted into a suitable cloning vector. Many vector host systems known in the art can be used. Vectors that can be used include, but are not limited to, plasmids or modified viruses. However, the vector system must be compatible with the host cell used. Such vectors include bacteriophages such as lambda derivatives, or PBR322 or pUC or CDM8 plasmids (Seed. B., 1987, Nature 329: 840).
-842) or derivatives such as, but not limited to. Recombinant molecules can be introduced into host cells by transformation, transfection, infection, electroporation, and the like.

Alternatively, the CR1 gene can be identified and isolated after insertion into a suitable cloning vector in a "shotgun" method. Prior to insertion into a cloning vector, the CR1 gene can be enhanced, for example, by fractionation by molecular weight.

Insertion of the CR1 gene into a cloning vector that can be used for transformation, transfection, or infection of a suitable host cell results in the production of multiple copies of the gene sequence. In a specific embodiment, a CDM8 vector that can be used to achieve expression in a mammalian host cell can be used as a cloning vector. For example, insertion into a cloning vector can be performed by ligating the DNA fragment to a cloning vector having complementary cohesive ends. However, if
If the complementarity restriction site used to cut the DNA into fragments is not present in the cloning vector, DN
The end of the A molecule can be modified enzymatically. Such modifications include the further digestion of the single-stranded DNA ends or the generation of blunt ends by filling in the single-stranded ends, so that these ends can be ligated as blunt ends. Alternatively, any desired site can be created by attaching a nucleotide sequence (linker) to the DNA terminus. These linked linkers can consist of specific chemically synthesized oligonucleotides encoding restriction endonuclease recognition sequences. Alternatively, the truncated vector and the CR1 gene can be modified by homopolymeric tiling.

The cloned CR1 gene can be identified in many ways, based on the properties of the DNA itself, or alternatively, on the physical, immunological, or functional properties of the protein it encodes. For example, DNA itself can be detected by plaque or colony nucleic acid hybridization to labeled probes (Benton, W. and Davi
s, R., 1977, Science 196: 180; Grunstein, M. and Hognes
s, D., 1975, Proc. Natl. Acad. Sci. USA 72: 3961). The presence of the CR1 gene can also be detected by assays based on the nature of the expressed product. For example, C
Electrophoretic migration, isoelectric focusing, proteolytic digestion maps, C3b and / or C4b and / or immune complex binding activity, similar or identical to that known for R1 A cDNA clone producing a protein having a body regulating activity, an effect on phagocytosis or immunostimulation, or an antigenicity, or a DNA clone hybrid-selecting a suitable mRNA can be selected. If an antibody against CR1 can be used, the CR1 protein can be identified by binding of the labeled antibody to a putative CR1 synthetic clone by an ELISA (enzyme linked immunosorbent assay) type method.

In a specific embodiment, the isolated CR1 gene, cDN
Multiple copies of the gene can be generated by transforming a host cell with A, or a recombinant DNA molecule incorporating a synthetic DNA sequence. Thus, it is possible to obtain the gene in large quantities by growing the transformant, isolating the recombinant DNA molecule from the transformant and, if necessary, recovering the inserted gene from the isolated recombinant DNA. it can.

In a specific embodiment, the CR1 cDNA clone of the CDM8 vector can be transfected into COS (monkey kidney) cells for large-scale expression under the control of a cytomegalovirus promoter (see 6.1.8 below). Section).

If the ultimate purpose is to insert the gene into a viral expression vector such as vaccinia virus or adenovirus, the recombinant DNA molecule incorporating the CR1 gene should be inserted such that the flanking side of the gene is flanked by viral sequences. Can be modified, thereby allowing recombination in the genome in cells infected with the virus, so that the gene can be inserted into the viral genome.

After identifying, growing, and harvesting the CR1 DNA-containing clone, the DNA insert can be characterized as described in Section 5.4.1 below.

Once the gene structure of the CR1 gene is known, it can be manipulated for optimal use in the present invention. For example, to increase protein expression,
Promoter DNA can be ligated 5 'to the CR1 coding sequence in addition to or in place of the native promoter. Expression vectors expressing CR1 deletion mutants are also made to provide for expression of specific fragments of the CR1 sequence (see Examples section below). In particular embodiments, the fragment encodes a required C3b and / or C4b binding function (see Section 9 below), eg, a fragment of the CR1 protein that exhibits LHR-A for C4b binding or LHR-C for C3b binding. Deletion mutants can be constructed. In a preferred embodiment, a CR having a deletion in the transmembrane region
An expression vector encoding one molecule can be used to produce a soluble CR1 molecule (see Examples 11-14 below). Many operations are possible, and they are within the scope of the present invention.

5.2. Expression of the CR1 Gene The nucleotide sequence encoding the CR1 protein (FIG. 1) or a portion thereof is inserted into a suitable expression vector, ie, a vector containing the necessary elements for the transcription and translation of the inserted protein coding sequence. be able to. Necessary transcription and translation signals are in the original CR
It can also be supplied from one gene and / or its neighboring regions. Various host vector systems are available for expressing the protein coding sequence. These include viral (eg, vaccinia virus, adenovirus, etc.) infected mammalian cell lines; virus (eg, baculovirus) infected insect cell lines; microorganisms such as yeast containing yeast vectors, or bacteriophage DNA, plasmid DNA, Or a bacterium transformed with a cosmid DNA, but is not limited thereto. The expression elements of these vectors vary in their strength and specificities. Depending on the host-vector system used, any one of a number of suitable transcription and translation elements can be used. For example, when cloning in mammalian cell lines, it is possible to use promoters isolated from the mammalian cell genome or promoters isolated from viruses that grow in mammalian cells, such as adenovirus, simian virus 40, and cytomegalovirus. it can. To cause transcription of the inserted sequence, the recombinant DNA
Alternatively, a promoter created by a synthetic technique can be used.

Specific initiation signals are also required for efficient translation of the inserted protein coding sequence. These signals include the ATG start codon and adjacent sequences. If the complete CR1 gene, including its own initiation codon and adjacent sequences, is inserted into the appropriate expression vector, no additional translation control signals need be added. However, if only a portion of the CR1 coding sequence has been inserted, an exogenous translational control signal including the ATG start codon must be provided. In addition, the initiation codon must be in reading frame with the protein coding sequence to ensure translation of the entire insert. Such exogenous translational control signals and initiation codons can be of various origins, both natural and synthetic.

To construct an expression vector containing a chimeric gene comprising appropriate transcription / translation control signals and a protein coding sequence, any of the methods described above for the insertion of a DNA fragment into a vector can be used. Such methods may include in vitro recombinant DNA and synthetic techniques, and in vivo recombination (genetic recombination).

In a specific embodiment, a soluble CR1 molecule is expressed. Such a soluble molecule contains D, which encodes the CR1 transmembrane region.
It can be produced by using recombinant DNA techniques that delete the NA sequence (see paragraphs 11-14 below). As shown below, soluble CR1
The ability to express the molecule is not limited to any genetic modification of the CR1 nucleic acid sequence. Thus, a soluble CR1 construct is obtained as long as the nucleic acid sequence encoding a substantial portion of the CR1 transmembrane region is deleted.

An expression vector containing a CR1 gene insert can be identified by three general methods. (A) DNA-DNA hybridization, (b) the presence or absence of "marker" gene function, and (c) expression of the inserted sequence. In the first method, the presence of the foreign gene inserted into the expression vector can be detected by DNA-DNA hybridization using a probe containing a sequence homologous to the inserted CR1 gene. In the second method, the recombinant vector / host system is used to define a particular "marker" gene function (eg, thymidine kinase activity, antibiotic resistance, transformation phenotype, etc.) caused by insertion of a foreign gene into the vector. Confirmation and selection can be made on the basis of the presence or absence of occlusion formation in baculovirus). For example, if the CR1 gene is inserted within the marker gene sequence of the vector, a recombinant containing the CR1 insert can be identified by a lack of marker gene function. In a third method, by assaying the foreign gene product expressed by the recombinant,
The recombinant expression vector can be confirmed. Such assays can be based on the physical, immunological, or functional properties of the gene product.

Once a particular recombinant DNA molecule has been identified and isolated, several methods known in the art can be used for its propagation. Once suitable host systems and growth conditions are established, recombinant expression vectors can be propagated and prepared in large quantities.

In certain embodiments detailed in the examples of the present invention,
A CDM8 vector with a CR1 cDNA insert can be transfected into COS cells. CR1 in this cell
The cDNA insert is expressed to produce the CR1 protein.
In another specific embodiment, detailed in the Examples section below, a CDM8 vector having a CR1 cDNA insert corresponding to a portion of the CR1 coding region is transfected into COS cells, where CR1 or a fragment is expressed. Further according to another example below, the cleaved soluble CR
One molecule is expressed in mammalian cells by using an expression vector such as the pTCS vector described in section 11.3.1. As already described, expression vectors that can be used include, but are not limited to, the following vectors or derivatives thereof, to name a few. Human or animal viruses such as vaccinia virus or adenovirus; insect viruses such as baculovirus; yeast vectors; bacteriophage vectors (eg, lambda), and plasmid and cosmid DNA vectors.

In addition, a host cell strain may be chosen which modulates the expression of the inserted sequences, or modifies and processes the chimeric gene product in the specific desired fashion. Expression from that type of promoter can be increased in the presence of a particular inducer. Therefore, genetically engineered
CR1 protein expression can be controlled. Furthermore, different host cells have characteristic and specific mechanisms for translation and post-translational protein processing and modification. Appropriate cell lines or host systems can be chosen to ensure the desired modifications and processing of the heterologous expressed protein. For example, expression in a bacterial system can be used to produce an unglucosylated CR1 protein having the deduced amino acid sequence of FIG. Expression in yeast is another embodiment that produces a glucosylated product.
To ensure “natural” glycosylation of proteins, mammalian COS cells can be used. In addition, various vector / host expression systems can perform processing reactions such as proteolytic cleavage to varying degrees.
Many such processed CR1 proteins can be produced, and are within the scope of the present invention. In a preferred embodiment of the present invention, the soluble CR
Large-scale production of one molecule is performed as described in Section 12.1 et seq.

5.3. Identification and Purification of Expressed Gene Product Once a recombinant expressing the CR1 gene has been identified, the gene product must be analyzed. This can be done by assays based on the physical, immunological, or functional properties of the product.

CR1 protein may be obtained by standard methods, including chromatography (eg, ion exchange, affinity and sizing column chromatography, high pressure liquid chromatography), centrifugation, differences in solubility, or any other standard for protein purification. Can be isolated and purified by conventional techniques.

In a preferred aspect of the invention, as detailed in the examples below, large amounts of soluble CR1 can be purified by a method using HPLC (see 12.2 et seq.). Purified CR1 as described below
Large-scale production of can be achieved by using expression methods that produce soluble CR1 as the starting material, thus eliminating the need to solubilize membrane-bound CR1 with detergent. The reduction of fetal calf serum concentrations in bioreactor cultures and / or the use of other cultures in these cultures has made it necessary to remove high levels of foreign proteins from the initial material containing soluble CR1 during subsequent purifications. Exclude. Anion exchange HPLC, after either cation HPLC or a combination of cation HPLC, can be used for purification in this preferred aspect. In this way, substantially pure soluble CR1 can be achieved in large quantities, in only one or two steps.

Further, once the CR1 protein produced by the recombinant is identified, the amino acid sequence of this protein can be deduced from the nucleotide sequence of the chimeric gene contained in the recombinant. As a result, the protein can be synthesized by standard chemical methods known in the art (eg, Hunkapiller, M. et al., 1984, Nat.
ure310: 105-111).

In a particular embodiment of the present invention, such a CR1 protein, whether produced by recombinant DNA techniques or by chemical synthesis,
Including, but not limited to, proteins containing substantially all or part of the amino acid sequence as the primary amino acid sequence as shown in FIG. The amino acid sequence also includes altered sequences in which functionally equivalent amino acid residues have been substituted for residues in the sequence to produce a silent change. For example, one or more amino acid residues in the sequence can be replaced by another amino acid of similar polarity and having a functionally equivalent effect, resulting in a silent change.
Also, non-conservative substitutions will result in a functionally equivalent protein.

In one embodiment, the replacement of an amino acid within the CR1 sequence can be selected from other amino acids in the class to which the amino acid belongs. For example, non-polar (hydrophobic)
Amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. Polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine, and glutamine. Positively charged (basic) amino acids include arginine, lysine, and histidine. Negatively charged (acidic) amino acids include aspartic acid and glutamic acid. CR1 proteins specifically modified during or after translation, for example, by glycosylation, proteolytic cleavage, etc., are also within the scope of the invention.

In the examples of the invention detailed below, the cloned recombinants expressed by the transfected cells
CR1 could not be distinguished from the erythrocyte allotype by SDS-PAGE (FIG. 14) and could mediate the binding of sheep erythrocytes carrying either C4b or C3b, demonstrating the ligand specificity of CR1. Reproduction (13th
Figure), and demonstrating the Factor I cofactor function for alpha polypeptide cleavage of C3 (ma)
(Figure 15).

5.4. Structure of CR1 gene and protein The structure of CR1 gene and protein can be analyzed by various methods known in the art, including but not limited to the methods described below.

5.4.1. Gene analysis The cloned DNA or cDNA corresponding to the CR1 gene was analyzed by Southern hybridization (Southern, EM, 1975, J. Mo.
l. Biol. 98: 503-517), Northern hybridization (e.g., Freeman et al., 1983, Proc. Natl. Acad. Sci. U.
SA80: 4094-4098), restriction endonuclease mapping (Maniatis, T., 1982, Molecular Cloning, A Lab)
oratory Manual, Cold Spring Harbor Laboratory, Cold
Spring Harbor, New York), and DNA sequence analysis, including, but not limited to. To ensure the detection of nucleic acids having the desired degree of relevance to the specific CR1 probe used,
The stringency of hybridization conditions for both Southern and Northern hybridizations can be adjusted.

Restriction endonuclease mapping can be used to roughly determine the gene structure of the CR1 gene. In certain embodiments, restriction enzyme digestion can be used to obtain the restriction map shown in FIG. 2 below. The restriction map obtained by restriction endonuclease cleavage can be confirmed by DNA sequence analysis.

DNA sequence analysis can be performed by any method known in the art. Maxa (Maxa
m) and the method of Gilbert (1980, Mech. En
zymol. 65: 499-560), the Sanger dideoxy method (Sange
r, F. et al., 1977, Proc. Natl. Acad. Sci. USA 74: 5463),
Or an automated DNA sequenator (eg, Applied B
iosystems, Foster City, CA). The cDNA sequence of the CR1 gene substantially includes the sequence shown in FIG. 1 and detailed in sections 6 and 7 below.

5.4.2. Protein Analysis The amino acid sequence of the CR1 protein can be derived by deduction from the DNA sequence, or alternatively by direct sequencing of the protein, for example using an automated amino acid sequencer. The amino acid sequence of a representative of the CR1 protein contains substantially the sequence shown in FIG. 1 and detailed in Section 6 below. As described below, all of the F allotype CR1 coding sequence has been cloned and, after cleavage of a 41 amino acid signal peptide, the mature receptor contains 1998 amino acids encompassing the 1930 ectodomain forming 30 SCRs. Including.

Of the 30 SCRs, 28 were LHR-A, -B, -C and -D (10th
Figure), comprising a single transmembrane domain of 25 amino acids and a relatively short cytoplasmic domain of 43 amino acids.

Among C3 / C4 binding proteins containing multiple SCRs, CR1 is LHR
Is unique in having a group of SCRs that make up C
By comparing the four LHRs of R1, each of the four SCR, a,
It is clear that the compound is composed of b, c and d species (FIG. 19). For example, SCR-1 and-of LHR-A
2 are the first two S of LHR-B, -C and -D.
It is 62%, 62% and 57% identical to CR, respectively. However, SCR-3 to SCR-7 differ from the corresponding SCR of LHR-B only at a single position,
And -4 differ from those of LHR-C only in three positions (FIG. 10). Therefore, the "a" species SCR of LHR-A
Are also present in LHR-B and -C. The first two SCRs of LHR-B, unlike those of LHR-A,
It is 99% identical to the corresponding SCR of C, so that LHR-B and -C share a "b" species SCR at these positions. LHR
The fifth, sixth and seventh SCRs of -C are 77% identical to the "a" species SCRs of LHR-A and -B at these positions and are considered "c" species SCRs. The first through fourth SCRs of LHR-D are relatively unique and are of the "d" species, while the fifth through seventh
Up to about 93% of the "c" species found in LHR-C
Are identical.

The CR1 protein sequence can be further characterized by hydrophilicity analysis (Hopp, T. and Woods, K., 1981, p.
roc.Natl.Acad.Sci.USA78: 3824). The hydrophilicity profile can be used to identify the hydrophobic and hydrophilic regions of the CR1 protein and the corresponding regions of the gene sequence encoding such regions. CO1 of CR1 protein
The terminal hydrophilicity profile is shown in FIG.

In order to identify the predicted region of CR1 where a specific secondary structure is assumed, secondary structure analysis (Chou, P. and Fasman,
G., 1974, Biochemistry 13: 222).

Other structural analysis methods can be used. These include X
Line crystallography (Engstom, A., 1974, Biochem. Exp. Biol. 11: 7-1)
3) and computer modeling (Fletterick, R. and Zoller, M. (eds.), 1986, Computer Graphics and Mol
ecular Modeling.in Current Communications in Molec
ular Biology, Cold Spring Harbor Laboratory, Cold Sp
ring Harbor, New York).

5.5. CR1 Related Derivatives, Analogs, and Peptides The production and use of derivatives, analogs, and peptides related to CR1 are also contemplated and are within the scope of the present invention. Such derivatives, analogs or peptides having the desired immunogenicity or antigenicity can be used for immunization, for therapeutic purposes, etc., for example in immunoassays. Such a molecule that retains or otherwise inhibits a desired CR1 property, such as, for example, binding of C3d or C4d, modulating complement activity, or promoting immune stimulation or phagocytosis, is an inducer of each such property. Or as an inhibitor.

The CR1-related derivatives, analogs and peptides of the present invention can be produced by various methods known in the art. The operations leading to such production can be performed at the gene or protein level. For example, a cloned CR1 gene can be modified in any of a number of ways known in the art (Ma
niatis, T. et al., 1982 Molecular Cloning, A Laboratory Ma
nnal, Cold Spring Harbor Laboratory, Cold Spring Har
bor, New York). The CR1 sequence can be cleaved in vitro with restriction endonucleases at appropriate sites and, if desired, further enzymatically modified, isolated and ligated (see Section 8 below). CR1-related derivatives, analogs,
Or in the production of a gene encoding a peptide,
Care should be taken to ensure that the modified gene has the same translational reading frame as CR1 in the gene region encoding the desired CR1-specific activity, without interruption by the translation termination signal It is. In a particular embodiment, a nucleic acid sequence can be generated that encodes the fusion protein and consists of a molecule comprising a portion of the CR1 sequence plus a non-CR1 sequence.

In addition, the CR1 gene may be mutated in vitro or in vivo to generate and / or disrupt translation, initiation, and / or termination sequences, or to make changes in the coding region, and / or to generate new restriction endonucleases. Sites can be formed or existing sites can be disrupted to further facilitate in vitro modification. Any method known in the art for mutagenesis can be used, such as in vitro site-directed mutagenesis (Hutchinson, C. et al., 1978, J. Bio.
l. Chem. 253: 6551), TAB ™ linker (Pharmaci
use of a), and the like, but are not limited thereto.

Manipulation of the CR1 sequence can also be performed at the protein level. Any of a number of chemical modifications can be made by known techniques, including cyanogen bromide, trypsin, chymotrypsin, pahain, V8 protease, NaBH
Specific chemical degradation by ₄ ; acetylation, formylation,
Oxidation, reduction; metabolic synthesis in the presence of tunica myinsin; and the like.

In addition, CR1-related analogs and peptides can be chemically synthesized. For example, a peptide corresponding to a portion of CR1 that mediates the desired function of gene expression (eg, C3b and / or C4b binding, immunostimulation, complement regulation, etc.) can be expressed as D
It can be synthesized using an NA synthesizer.

A particular change in the nucleotide sequence of CR1 is multiple LHR-B
This is done by a recombinant DNA method that results in a sequence encoding a protein having the sequence. Such a binding change alters the degree of C3b binding.

5.6. Use of CR1 5.6.1. Assays and Diagnostics The CR1 protein, its analogs, derivatives and subsequences, and anti-CR1 antibodies have diagnostic use. Desired C
Molecules of the invention that exhibit R1 properties or functions can be used to assay such properties or functions. For example, C3b and / or C4b alone and / or in complex form
The CR1 protein or a fragment thereof exhibiting the binding of can be used in an assay to determine the amount of such a substance in a sample, for example, a body fluid of a patient.

In a specific embodiment, C3b (eg, Table II below,
9), full-length CR1 or CR1 deletion mutants expressed on the cell surface capable of binding to iC3b or C4b (see, eg, Table II) (see, eg, Section 8 below). Can be used in assays that measure the levels of C3b, iC3b, or C4b, respectively, in a sample. In other embodiments, constructed by recombinant DNA technology,
The CR1 protein or a fragment thereof lacking a transmembrane sequence is thus secreted and used.

In particular embodiments, such measurements of C3b and / or C4b are reliable as indicators of complement function and are useful in diagnosing inflammatory and immune system disorders. Such disorders include tissue damage from burns or myocardial infarction-induced injury, adult respiratory distress syndrome (shock lung), autoimmune disorders such as rheumatoid arthritis, systemic lupus erythematosus, and unwanted or inappropriate complement function. Including, but not limited to, other diseases or disorders associated with (e.g., Miescher, P. et al.
A. and Muller-Eberhard, HJ, (eds.), 1979, Text Boo
k of Immunopathology, 2d Ed., Vols.I and II, Grune an
d Stratton, New York; Sandberg, AL, 1981, in Cellular
Functions in Immunity and Inflammation, Oppenheim,
JJ et al. (Eds.), Elsevier / North Holland, New York, p.3
73; Conrow, RB et al., 1980, J. Med.Chem. 23: 242; Regal, J.
F. and Pickering, RH, 1983, Int. J. Immunopharmacol.
5:71; Jacobs, HS, 1980, Arch. Pathol. Lab. 104: 617).

The CR1 protein containing the epitope and fragments thereof have use in assays, including but not limited to immunoassays. Immunoassays that can be used include:
Radioimmunoassay, ELISA (enzyme-linked immunosorbent assay), "sandwich" immunoassay, sedimentation reaction, gel diffusion sedimentation reaction, immunodiffusion assay, agglutination assay, complement fixation assay, immunoradiometric assay, fluorescence to name a few. Competitive and non-competitive assay systems using techniques such as immunoassays, protein A immunoassays, and immunoelectrophoresis assays are included, but are not limited to.

The CR1 gene and related nucleic acid sequences and subsequences can be used in hybridization assays.
Such hybridization assays monitor inflammatory or immune responses associated with CR1 expression, diagnose specific disease states associated with altered CR1 expression, determine the CR1 allotype of a patient, and Can be used to detect the presence and / or expression of a modified gene (eg, CR2).

Kits for performing the assays used in the present invention are also provided.

5.6.2. Therapies CR1 proteins and fragments, derivatives, and analogs thereof are therapeutically useful for modulating CR1-mediated functions. Such functions include, but are not limited to, binding of C3b and / or C4d, alone or in complex, promotion of phagocytosis, complement regulation, immune stimulation, and the like. Dosages that exert an effect of the CR1 protein and related molecules of the invention have therapeutic value in many diseases or disorders associated with such functions, such as immune or inflammatory disorders (see, eg, section 5.6.1 above). Obstacles described in For example, full-length CR1 or a fragment thereof and the related molecule exhibiting the desired function may comprise complement component C3.
Factor I promoting irreversible inactivation of C4b or C4b
Due to its ability to act as a cofactor (Fearo
n, DT, 1979, Proc.Natl.Acad.Sci.USA76: 5867; Iida,
K. and Nussenzweig, V., 1981, J. Exp. Med. 153: 1138) and / or have therapeutic use in inhibiting complement by the ability to inhibit the C3 or C5 conversion of the alternative or major pathway .

In a specific embodiment of the invention, to encode a CR1 molecule that lacks a transmembrane region (eg, by deletion of the carboxy terminus from arginine encoded by most C-terminal SCRs) resulting in production of a soluble CR1 fragment. Expression vectors can be constructed. In one embodiment, such fragments may be used alone or in a complex.
The ability to bind C3b and / or C4b can be retained. In a specific embodiment, such a soluble CR1 protein no longer exhibits Factor I cofactor function. The soluble CR1 product is administered to the patient in vivo,
Soluble CR1 is responsible for CR1 on the intrinsic cell surface.
It overcomes the binding competition of C3b and / or C4b, thus blocking cell surface CR1 factor I cofactor function and increasing complement function.

After C3b covalently attaches to particles and soluble immune complexes, it is proteolytically processed to form iC3b and C3
Inactivation of C3b to dg has two biological consequences.
Prevention of excessive activity of complement tissue via the proliferative pathway, and formation of ligands associated with receptors other than CR1. Since the iC3b fragment is unable to bind factor B, switching to this state blocks follow-on complement activity via the alternative pathway amplification circuit. However, iC3b can bind by CR1 and CR3, two complement receptors that mediate phagocytosis by myeloid mononuclear cells. Therefore, the major biological consequence of the conversion of C3b to iC3b is
Abort complement activity without interfering with CR1- and CR3-mediated clearance of the complex covered with. In contrast iC3b
The subsequent conversion from C3dg to C3dg only interacts with CR2 but
Create fragments that do not interact with R1 and CR3.
This situation limits the complement-dependent binding of C3dg-bearing complexes to CR2-expressing cell types, including B lymphocytes, follicular dendritic cells, and possibly dermal epithelial cells, and inhibits interaction with phagocytic cell types. Reduce or block action.
The biological consequence of the altered pattern of cellular involvement would be the target of the C3dg-loaded complex on cells involved in the induction phase of the immune response rather than on cells involved in clearance and degradation of particles and complexes. Therefore,
In addition to affecting the clearance process, CR1 molecules can be used therapeutically for CR2-bearing cell types involved in antigen presentation and antibody production.

In other embodiments, the agent binds to C3b or C4b and / or retains the ability to block C3 or C5 conversion of the alternative or major pathway, or a factor I
-The CR1 protein or a fragment thereof carrying the cofactor can be used to promote complement inactivation. In such embodiments, the CR1 protein or fragment is of value in the treatment of disorders involving unwanted or inappropriate complement function (eg, shock lung,
Tissue damage, autoimmune disorders, inflammatory diseases, etc. due to burns or bloody heart abnormalities).

In particular embodiments, detailed in the examples of paragraphs 11-14 below, for example, the ability to inhibit the complement-mediated hemolysis of the main pathway, production of the main pathway C5a, production of the main pathway C3a or neutrophil oxidative destruction in vitro. A soluble CR1 molecule that retains the desired functional activity as expressed by is expressed. In particular embodiments, such soluble CR1 molecules can be used to reduce inflammation and its deleterious effects, or to reduce the size of myocardial infarction, prevent reperfusion injury, and the like. . Such CR1 molecules useful for in vivo therapy are tested in a variety of model systems known in the art, including, but not limited to, the reverse passive Arthus reaction (see Section 14.1) and the rat myocardial infarction model (see Section 14.3). Is done.

In another embodiment of the present invention, the desired CR1
Fragments or analogs of CR1 that exhibit a property or function, or derivatives thereof, can be used to prevent or treat a disease or disorder associated with that function.

Various release systems are known and they can be used to release CR1 and related molecules. For example, encapsulation in liposomes, expression by hematopoietic stem cells in gene therapy, and the like. Other methods for introduction include, but are not limited to, intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous and intranasal and oral routes.

The present invention also provides a pharmaceutical composition. Such a composition is
Comprising a therapeutically effective amount of the CR1 protein, or analog, derivative, or fragment thereof, and a pharmaceutically acceptable carrier. Such carriers include, but are not limited to, saline, buffered saline, dextrose and water.

5.6.3. Combination therapy In a further aspect, there is provided a method of treating thrombotic symptoms in humans and animals, particularly acute myocardial infarction. The method comprises administering to a human or animal in need an effective amount of a soluble CR1 protein and an effective amount of a thrombolytic agent in accordance with the present invention.

The present invention also provides the use of a soluble CR1 protein and a thrombolytic agent in the manufacture of a medicament for the treatment of thrombotic conditions in humans and animals.

In the above methods, the compounds can be administered by any convenient route, for example, by infusion or bolus injection, or can be administered sequentially or all at once. When sequentially administering the soluble protein and the thrombolytic agent of the present invention, soluble CR1
The protein can be administered either before or after the thrombolytic agent. When the soluble CR1 protein and the thrombolytic agent are administered together, they are preferably provided in the form of a pharmaceutical composition comprising both agents. Thus, in a further aspect of the invention there is provided a pharmaceutical composition comprising a soluble CR1 protein and a thrombolytic agent together with a pharmaceutically acceptable carrier.

In a preferred embodiment, the compositions are formulated according to conventional methods as pharmaceutical compositions suitable for intravenous administration to humans.

Typically, compositions for intravenous administration are solutions in sterile isotonic aqueous buffer. Where necessary, the composition will also include an availability agent and a local anesthetic such as lignocaine to ease pain at the site of the injection. Generally, the components are separately dispersed in unit dose form as a dry lyophilized powder or a water-free concentrate in a tightly sealed container such as an ampoule or sachette indicating the amount of active agent in the active unit. Or, they will be supplied as a mixture. Where the composition is to be administered by infusion, it is dispensed with an infusion bottle containing sterile pharmaceutical grade "water for injection" or saline. Where the composition is administered by injection, an ampoule of sterile water for injection or saline is provided and the components are mixed prior to administration. Pharmaceutical packs comprising one or more containers filled with one or more components of the pharmaceutical composition are also within the scope of the invention.

The amount of substance administered and the ratio of thrombolytic agent to CR1 protein will depend on the severity of the thromboembolic symptoms and the location and size of the clot. The exact dose and mode of administration to be used must be determined circumstanceally by the treating physician, taking into account the nature of the illness, either or not. However, in general, patients being treated with thrombus will generally take a dose of 0.5-50 mg of complement inhibitor (soluble CR1 component) per standard dose of thrombolytic agent.

Particular thrombolytic agents used in the combination therapy described above are fibrinolytic enzymes including plasminogen activator.

The term plasminogen activator includes, but is not limited to, streptokinase, human tissue plasminogen activator (t-PA) and urokinase (u-PA) (both single and two chain forms). Such enzymes can be obtained from natural sources or tissues or by recombinant DNA methods in which a heterologous host organism, such as a bacterium, yeast, fungus, or animal cell, expresses a gene specific for the enzyme. The term also includes: (a) at least one of the chains of the hybrid protein is derived from a fibrinolytically active protease such that the hybrid protein has a catalytic site obligatory for fibrinolytic activity that is optionally blocked with a removable blocking group. EP-A-0155387 and EP-, which disclose a fibrinolytically active hybrid protein comprising one chain of a double-stranded protease bound to one chain of a different double-stranded protease. A protein disclosed in A-297882, (b) a protein conjugate such as urokinase conjugated to reversible block plasmin disclosed in EP-A-0152736, (c) an enzyme catalyzing fibrinolytic activity A reversible linking group, such as urokinase, whose site is reversibly linked to the active center of human plasmin More, EP-A-01 that was blocked by a human protein attached thereto
A derivative of the fibrinolytic enzyme disclosed in 55388, (d) a conjugate comprising a fibrinolytic enzyme linked to a water-soluble polymer by a reversible linking group disclosed in EP-A-0183503, and (e) ) EP-A-0 like des (cys ₅₁ -asp ₈₇ ) t-PA
201153, EP-A-0207589, WO (PCT Gazette) -8604351, EP-
0041766, EP-0213794, EP-0199574, EP-A-020334, EP
−A−0241208, EP−A−0241209, EP−A−0241210, EP−
A-0233013, EP-A-290118, EP-A-292326, EP-A-
0213794, EP-A-0231883, WO8704722, EP-A-022836,
EP-A-0234051, EP-A-0253582, EP-A-0253241, WO
-8604351, EP-A-0236040, EP-A-0200451, EP-A-
0238304, EP-A-0225286, DE (West German publication) -353717
6, EP-A-0236289, WO-8601538, EP-227462, AU (Australian publication) -8661804, WO-8703906 and EP-019.
And engineered derivatives, including the muteins of the naturally occurring plasminogen activator disclosed in 9574.

In a particular aspect of the invention, the plasminogen activator is the t-PA cleavage site between residues 275 and 276 of t-PA and cysteine residue 264 or 158 and 159 of u-PA.
Cleavage site between the residue and cysteine residue 148, via the amino acid sequence comprising each, t-PA or u-PA
A hybrid molecule as described in EP-A-0297882, comprising five kringle domains of plasminogen bound to the β-chain of PA.

Examples of such hybrids are the 1 and 2 chain variants, ly
s ₇₈ and glu ₁ variants, and includes a mixture plasminogen 1-544 / t-PA262-527, 1 and 2 chain variants, lys ₇₈ and glu ₁ variants, and plasminogen including mixtures thereof Gen 1-544 / t-P
A262-527 (arg ₂₇₅ gln) Plasminogen 1-541 / t-P including 1 and 2 chain variants, lys ₇₈ and glu ₁ variants, and mixtures thereof
A262-527 1 and 2 chain variants, gly _-3, ser ₁ and val ₄ variants, and t-PA1-50 / t-PA88 including mixtures thereof
-91 / pro-gly-ser / plasminogen 84-544 / t-PA26
T-PA1-91 / pro, including 2-527 1 and 2 chain variants, gly- ₃ , ser ₁ and val ₄ variants, and mixtures thereof
Plasminogen 1-546 / u, including gly-ser / plasminogen 84-544 / t-PA262-527 or 1 and 2 chain variants, lys ₇₈ and glu ₁ variants, and mixtures thereof
-PA137-411.

In a preferred embodiment, the thrombolytic agent used in the combination therapy reversibly blocks in vivo fibrinolytic enzymes having the meaning set forth in Smith of US Patent No. 4,285,932. In other words, the catalytic site essential for fibrinolytic activity has a pseudo-first-order rate constant for hydrolysis at pH 7.
4. In vivo fibrinolytic enzymes blocked by groups that can be removed by hydrolysis at a rate such as in the range of 10 ^-8 sec ^-1 -10 ^-3 sec ^-1 at 37 ° C.

When the fibrinolytic enzyme is a plasminogen activator comprising the serine protease domain of t-PA or urokinase, examples of removable blocking groups are substituted with halogen atoms and one or more weak electron-requesting groups. Optically substituted with an attracting or electron donating group,
A 2-aminobenzoyl group at the 3 or 4 position,
The pseudo-first-order rate constant for the hydrolysis of the derivative is
0.05M sodium phosphate, 0.1M sodium chloride, about 1300
6.0 × 10 ⁻⁵ −4.0 × 10 ⁻⁴ s when measured at 37 ° C. in a buffer system consisting of 0.01% v / v detergent comprising polyoxyethylene sorbitan monooleate having a molecular weight of
It is in the range of ec ^-1 .

Preferably, the in vivo fibrinolytic enzyme that is reversibly blocked is two of streptokinase and plasminogen.
The component complex, most preferably referred to as APSAC after the trademark EMINASE (trade name anistreplase).
That is, anisoylated human plasminogen-streptokinase-activator complex. For example, JPMonk a
nd RCHeel, 1987, Drugs 34 : 25-49) in Beecham
US Patent No. 4,808, marketed by Group plc
405 is a p-anisoyl streptokinase / plasminogen complex without internal bond cleavage as described in 405.

In a preferred aspect, the soluble CR used in combination therapy
One component is pBSCR1c, pBSCR1s, pBM-CR1c, pBSCR1c / pTCSgpt
And the above pBSCR1c / pTCSgp encoded by a nucleic acid vector selected from the group consisting of pBSCR1s / PTCSgpt
It has been specially prepared from t (see section 12).

Specific thrombolysates used in combination therapy are as follows (example doses and methods of administration):

6. Example: Cloning and Sequencing of the Human C3b / C4b Receptor (CR1) In the examples detailed herein, the 5.5 kilobase pair (K
b) Illustrates the cloning and nucleotide sequence of the CR1 coding region of (Klickstein, LB et al., 1987, J. Exp. Med.
165: 1095-1112).

Ten overlapping CR1 cDNA clones spanning the 5.5 Kb range were isolated from a library of tonsils and sequenced in whole or in part. A single long reading frame was identified beginning at the 5 'end of these clones and continuing to a stop codon 4.7 kb downstream. This corresponds to 80% of the 6 kb calculated for the coding sequence of the F allotype of CR1. Three tandem, identical, long, 450 amino acid homologous repeats (LHR) were identified. By trypsin peptide sequence analysis, CR
The presence of a fourth LHR was demonstrated in one F allotype. Amino acid identity between LHRs was 7 between the first and third repeats.
0%, the first and NH ₂ of the second iteration - ranged up to 99% between the terminal 250 amino acids. Each LHR comprises seven short consensus repeats (SCRs) of 60-70 amino acids, which include, for example, complementary receptor type 2, factor B and H, C4 binding proteins, and other C3s such as C2. Similar to SCR of / C4 binding protein. Two additional SCRs bind the LHR to a single 25-amino acid membrane-bound domain. Thus, the F allotype of CR1 probably contains at least 30 SCRs, 23 of which have already been sequenced. Each SCR is expected to form a triple loop structure in which the four conserved 1 / 2-cystine form a disulfide bond. 30 SCRs in a row as a semirigid structure 1.140 from the plasma membrane
Angstrom could be facilitated by the interaction of CR1 with C3b and C4b located in the void between the immune complex and the microbial cell wall. The 43-residue COOH-terminal cytoplasmic domain contains a sequence of 6 amino acids, which is homologous to the sequence at the epidermal growth factor receptor that is phosphorylated by protein kinase C.

6.1. Materials and Methods 6.1.1. Isolation and Sequence of CR1 Tryptic Peptide CR1 was purified from washed human erythrocyte membranes by subsequent Matrix Red A and YZ-1 monoclonal antibody affinity chromatography (W
ong, WW et al., 1985, J. Immunol. Methods 82: 303). (Won
g, WW et al., 1985, Proc. Natl. Acad. Sci. USA 82: 7711)
Tryptic peptides were prepared and isolated by sequential gradient and isocratic reversed-phase HPLC (high performance liquid chromatography) as described in. 470A protein sequencer (Applied Biosys
tems, Inc., Foster City. CA), and analysis of each degradation cycle was performed using a 120 PTH-amino acid analyzer (Applied Biosystems, Inc.).

6.1.2. Isolation of cDNA and Genomic Clones Description (Wong, WW et al., 1985, Proc. Natl. Acad. Sci. US
A.82: 7711) and human tonsil poly (A) + RNA
A cDNA library was constructed from λgtll. By RNA blot hybridization, tonsil donors were homozygous for the F allele of CR1 (Id.). CDNA including fragment between 2 and 7 Kb before cloning step
Was selected on agarose gel. The initial recombination rate of this library was 4.5 × 10 ⁶ recombinants per 100 ng cDNA, and the library was transformed into Eschericia coli.
hia coli) Y1088 strain. The library was prepared using the CR1 probe, CR1-1 (ATCC No. 57330 (CR1-1 plasmid containing E. coli, 57331 (purified CR-1 DNA)) and CR1-2 (W
ong, WW et al., 1985, Proc. Natl. Acad. Sci. USA 82: 771.
1) (Maniatis, T. et al., 198).
2, Mole-cular Cloning, A Laboratory Manual, Cold Spr
ing Harbor Laboratory, Cold Spring Harbor, New Yor
k). These probes were previously radiolabeled by nick translation to a specific activity of 2-8 × 10 ⁸ cpm / μg. In 50% formamide, 5 × SSC (1 × SSC: 15 mM sodium citrate, 150 mM sodium chloride) at 43 ° C.
Hybridize and filter at 60 ° C with CR2 cD
Washed with 0.2 × SSC under conditions where no NA clones are detected (Weis, JJ et al., 1986, Proc. Natl. Acad. Sci. USA 83:
5639). After plaque purification of the positive clone twice, restriction mapping and DNA sequence analysis were performed.

15- produced by Sau3A I partial digestion of human leukocyte DNA
Using a 20 kb fragment, the genomic library was
3 was constructed. The initial recombination rate was 1.2 × 10 ⁶ and the library was amplified in E. coli P2392. This library was also screened using the cDNA probes CR1-1 and CR1-2 (Wong, WW et al., 1985, Proc. Natl. Acad. S.
ci.USA82: 7711).

6.1.3. DNA sequence analysis Restriction fragments of cDNA clones were subcloned into M13mp18 or M13mp19, and the dideoxynucleotide chain termination method (Sanger, F. et al., 1977, Proc. Natl. Aca.
d.Sci.USA74: 5463). Some clones were sequenced in whole or in part by an ordered deletion mutant initially created using exonuclease III (Henikoff, S., 1984, Gene 28: 351). Each region was sequenced double-stranded, and in most cases, each region was sequenced with an M13 subclone constructed from two independently isolated cDNA clones (FIG. 2). W
Sequence data was analyzed using the isconsin University Genetic Computer Group Package (Madison, WT).

6.2.Results 6.2.1.Nucleotide sequence of CR1 gene
Screening was performed using CR1-1 and CR1-2 probes obtained from NA clone, λT8 (Wong, WW et al., 198
5, Proc. Natl. Acad. Sci. USA 82: 7711). Fifteen positive phages were identified from 1.5 × 10 ⁸ recombinants, 13 of which corresponded to different clones. Restriction maps were prepared for the ten, and the whole or a part thereof was sequenced by the dideoxy chain termination method. CDNA clones were aligned based on the overlap of identical sequences (FIG. 2), and 5.5 kb
(Fig. 3). A single long reading frame was identified beginning at the 5 'end of the cDNA clone and extending 4.7 kb downstream of the stop codon. The coding sequence for CR1 in this library is expected to be 6 kb based on the estimated molecular weight of the non-glycosylated receptor of 220,000 daltons (Wong, WW et al., 1983, J. Clin. Invest. 72:68
Five). Thus, these clones cover about 80% of the putative coding sequence.

Clones T49.1 and T55.1 contain the coding sequence at their 5 'end. This indicates that other 5 'coding and non-coding sequences remain unidentified.
Clones with overlaps in the 3 'region, T8.2, T4
3.1 and T87.1 contain the transmembrane and cytoplasmic regions encoded by the same sequence in each clone. Most 3 '
Flanking clone T8.2 has no poly (A) sequence.
Contains a 7 bp untranslated region. Clone T6.1 and T55.1
Clone T8.3 has a 91 bp deletion of nucleotides 1,406-1,497, compared to the sequence found in clone T40.
One has a 9 bp deletion of nucleotides 1,498-1.507. These deletions occurred in regions having sequences homologous to the 5 'splice site, but such deletions may indicate a splicing error in the mRNA. clone
T49.1 and T55.1 are reading frame nucleotides
It has a 110 bp insertion between 147 and 148 (FIG. 3). This sequence did not hybridize to a tonsil poly (A) ⁺ RNA blot and had a 5 'splice site (Breat hnach, R. et al., 1978, Proc. Natl. Acad. Sc.
iUSA75: 4853) (FIG. 3), which is determined to be a part of an intron because it is adjacent to the cDNA sequence in the CR1 genomic clone and its reading frame is shifted. Clone T9.4 has a 0.88 kb intervening sequence at the 3 'end that does not hybridize to a block of tonsillar poly (A) ⁺ RNA.

6.2.2. Nucleotide and amino acid sequence analysis of CR1 Dot matrix analysis of the nucleotide sequence of CR1 (FIG. 3) showed two types of internal homology (FIG. 4). The first type of internal homology is represented by a thick solid line, which is three columns, in the same direction, with high homology 1.
Shows a 35 kb repeat. These nucleotide sequences encode the long homologous repeat (LHR) of CR1. The second type of repeat is represented by a dashed parallel line, which indicates a region of relatively low homology. Such sequences occur every 190-210 nucleotides and encode a short consensus repeat of CR1.

The amino acid sequence deduced from the cDNA sequence is shown in FIG. Three Ls named LHR-B, LHR-C and LHR-D
HRs are aligned to demonstrate their homology. LHR-B ranges from residue 1 to residue 438, LHR-C corresponds to residues 439-891, and LHR-D ranges from residues 892 to 1,341. LHR-C
Residues 451-694 are 99% identical to residues 1-244 of LHR-B, but only 61% identical to the corresponding residues of LHR-D. In contrast, residues 695-891 of LHR-C are 91% identical to residues 1,148-1,341 of LHR-D, but only 76% identical to the corresponding region of LHR-B. Therefore LHR−
C appears to be a hybrid comprising very homologous sequences in the first half of LHR-B and the second half of LHR-D.
Following the LHR are two SCRs without repeats, a 25 residue hydrophobic segment and a 43 amino acid COOH-terminal region with no sequence homology to the SCR (FIG. 5).

The 5 '1.3 kb CR1 coding sequence corresponds to the fourth LHR, LHR-A (see FIG. 1 above and section 7 below). This conclusion was supported by the analysis of the trypsin peptide of erythrocyte CR1. The 10 tryptic peptides have the same sequence as the amino acid sequence from the cDNA clone (Table I).

Each LHR comprises seven 60-70 amino acid SCRs, which characterize a group of C3 and C4 binding proteins (C4bp) (FIG. 6A). The greatest homology between the 23 SCRs of CR1 is
Observed by introducing space in the alignment (No.
6A). 29 out of an average of 65 residues in each repeat are conserved. There are six residues present in all SCRs. Four 1/2 cystine, which are in relatively similar positions, suggesting that each of them may be involved in an essential disulfide bond. And a second glycine after tryptophan and a second 1/2 cystine (FIG. 6A). Choose the sequence between the unchanged 1/2 cystine Cho
u and Fasman's algorithm (Chou, PY and Fasman, GD, 19
74, Biochemistry 13: 222), it was predicted that the possibility of β-turn formation was high and the possibility of α-helix formation was low by secondary structure analysis. Two CRs
Sequence analysis of one genomic clone, 2.38 (FIG. 6B) and 2.46, shows that SCR-14 (FIG. 6A) is encoded by a single exon and that the COOH-terminus of SCR-23 corresponds to the exon end To do so. Therefore, Factor B (Campbell, RD and Bentley, DR, 1985, immuno
87:19) and IL-2-R (Leonard, WJ et al., 1985, Sc.
ience230: 633) Separate exons may encode the CR1 SCR as indicated for the SCR.

The consensus sequence of the SCR of CR1 is compared with the SCRs of another group of members having such a characteristic structure (see FIG. 7).
Figure). These members are proteins having a C3 / C4 binding function, namely CR2 (Weis, JJ et al., 1986, Proc. Natl. A
cad.Sci.USA83: 5639), C4bp (Chung, LP et al., 1985,
Biochem, J. 230: 133), Factor H (Kristensen, T .;
1986, J. Immunol. 136: 3407), Factor B (Morle
y, BJ and Campbell, RD, 1984, EMBO J. 3: 153; Mole,
JE et al., 1984, J. Biol. Chem. 259: 3407), and C2 (Ben
tley, DR and Porter, RR, 1984, Proc.Natl.Acad.Sc
iUSA81: 1212; Gagnon, J., 1984, Philos.Trans.R.Soc.
Lond. B. Biol. Sci. 306: 301) as well as, for example, the interleukin-2 receptor (Leonard WJ et al., 1985, Sci.
_{ence230:. 633), β 2} - glycoprotein I (Lozier, J et al,
1984, Proc. Natl. Acad. Sci. USA 81: 3640), Clr (Leyt
us, SP et al., 1986, Biochemistry 25: 4855), haptoglobin alpha chain (Kurosky, A. et al., 1980, Proc. Natl. Acad. Sci. U.
SA77: 3388) and factor XIIIb (Ichinose, A. et al., 19).
86, Biochemistry 25: 4633), and proteins that are not known to have such a function.
The 1/2 cystine residue is invariant in the SCR of all proteins, except for haptoglobin, which lacks the third 1/2 cystine. Tryptophan is also immutable, beta ₂ - 2 pieces of repeats present in the fifth SCR and the XIII b Factor glycoprotein I is an exception. Other residues that are conserved but not present in all SCRs tend to cluster around 1/2 cystine. Factors B and C2 have only one free thiol group (Christie, DL and Gagnon, J., 1982, Biochem. J. 201:
555; Parkes, C. et al., 1983, Biochem. J. 213: 201), β ₂ −
In the SCR of glycoprotein I, the first 1/2 cystine is disulfide bonded to the third and the second to the fourth (Lozi
er, J. et al., 1984, Proc. Natl. Acad. Sci. USA 81: 3640).

The resulting amino acid sequence of CR1 has 17 potential sites for N-linked oligosaccharides, all of which are in the extracellular region (FIG. 6A). Difference in CR1 molecular weight synthesized in the presence and absence of tunicamycin (Lublin, DM
1986, J. Biol. Chem. 261: 5736) and analysis of glucosamine content (Sim, RB, 1985, Biochem. J. 232: 883) showed that only 6-8 N-linked complex oligosaccharides To indicate that not all potential sites are utilized. For example, the obtained amino acid sequence (fifth
Asparagine at residue 263 in FIG.
Table 2), which indicates no glycosylation at this site. In contrast, the unidentified amino acid of peptide 34b probably corresponds to a glycosylated asparagine at residue 395.

Only the non-repeated CR1 sequence identified in the 5.5 kb cDNA is present in the COOH-terminal region. Secondary structure analysis of this region identifies a putative membrane-associated segment consisting of a single 25 residue. This segment has a strong hydrophobic character and has a very high potential to form an α-helix (FIG. 5). Immediately following this sequence are four positively charged residues, a property unique to many membrane proteins. The putative cytoplasmic region of CR1 has 43 residues and contains a sequence of 6 amino acids, VHPRTL. This sequence is a protein kinase C phosphorylation site in the epidermal growth factor (EGF) receptor and erbB oncogene product, VRKR.
TL, (Hunter, T. et al., 19845, Nature 311: 48
0; Davis, RJ and Czech, MP, 1984, Proc. Natl. Acad. S
ci.USA82: 1974). There are no tyrosine residues in the cytoplasmic region of tonsil CR1.

6.3 Discussion Approximately 80% of the primary structure of the F allotype of CR1 was obtained by sequencing overlapping cDNA clones. The most distinctive structural feature of CR1 observed in this analysis is the presence of tandem, co-directed, 450 amino acid LHRs, with four LHRs representing an estimated polypeptide chain length of 2,000 residues. Is expected to be present in the F allotype of CR1 (W
ong, WW et al., 1983, J. Clin. Invest. 72: 685; Sim, RB, 198.
5, Biochem. J.232: 883). While three LHRs were cloned and sequenced, evidence for the presence of a fourth LHR was provided by tryptic peptide analysis.
Each LHR is composed of seven SCRs, which are the basic structural units of other C3 / C4 binding proteins. Conservation of four 1/2 cystine per SCR, first and third, and second and fourth 1/2
Cystine is probably involved in disulfide bonds (Lozier, J. et al., 1984, Proc. Natl. Acad. Sci. USA 81:36
40), and the presence of conserved amino acids such as proline, glycine, and asparagine, which are common in β-turns (Rose, GD et al., 1985, ADv. Protein Chem. 37: 1).
Lead to the proposal that the SCR forms a triple loop structure maintained by disulfide bonds (FIG. 8). 1/2
Such a role for cystine may be due to trypsinized CR1 under mild conditions (Sim, RB, 1985, Biochem. J. 232: 8
83) and Factor H (Sim, RB and DiScipio, R.
G., 1982, Biochem. J. 205: 285), when analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) under non-reducing conditions, migrates similarly to the untreated molecule, and after reduction, a number of trypsin fragments. Supported by the discovery that it moves as well.

Such a series of tandemly repeated SCRs is expected to form the submitted extended structure (FIG. 8) for Factor H and for each subunit of human C4bp (Sim, RB and DiScipiD, RG, 1982, Biochem,
J. 205: 285; Whaley, K. and Ruddy, S., 1976, J. Exp.Med. 1
44: 1147; Daahlback, B. et al., 1983, Proc. Natl. Acad. Sci. U.
SA80: 3461). Electron microscopy studies of the C4bp subunit showed a size of 300 × 30 angstroms (Dahlback, B. et al., 1983, Proc. Natl. Acad. Sc.
iUSA80: 3461). Since each subunit is composed of eight SCRs (Chung, LP et al., 1985, Biochem. J. 230: 133), one SCR is calculated to be 38 × 30 Å. Given that the SCRs of CR1 have similar sizes and that the F allotype has 30 SCRs, the receptor could extend as much as 1,140 angstroms from the cell membrane. Ferritin-labeled antibody bound to CR1 on neutrophils was often 500 Å from the outer lobe of the plasma membrane (Abrahamson, DR and Fearon, D. et al.
The early expression of T.1983, Lab. Invest. 48: 162) is consistent with such a CR1 structure expectation. Such an extended structure of CR1 is composed of cells bearing receptors and C3 covalently bound to relatively inaccessible sites in the immune complex and to the surface of microbial cells.
Will facilitate interaction with b.

The finding that the SCR is the major and probably only extracytoplasmic element of CR1 suggests a close relationship between the receptor and the two plasma proteins, factor H and C4bp, which are solely or predominantly composed of SCRs (Chung, LP et al., 1985, Bioch.
em.J. 230: 133; Kristensen, T. et al., 1986, J. Immunol. 136: 3
407). CR1 was first isolated as a erythrocyte membrane protein with factor H-like activity after detergent solubilization (Fe
aron, DT, 1979, Proc.Natl.Acad.Sci.USA76: 586
7) Then, when present on the plasma membrane, factor H
And C4bp regulatory function (Iida, K. and Nussenzweig, V., 1981, J. Exp. Med. 153: 1
138). Analysis of the CR1, factor H, and C4bp structural polymorphism genes indicated that the genes encoding these three proteins were linked (de Cordoba, R. et al., 1985, J. Exp. .Med. 161: 1189), a CR of this linkage group and structurally related receptors.
The two loci have been shown to be on the long arm of chromosome 1, band q32, by in situ hybridization in vivo and by analysis of somatic cell hybrids (Weis,
JH et al., 1987, J. Immunol. 138: 312). Prior to this study, the only evidence for a structural relationship between these proteins was the significant similarity in their amino acid composition (Wong, WW et al., 1985, J. Immunol. Methods 82: 303). Thus, the discovery of at least 23 of these SCRs in CR1 suggests a structural relationship between the receptor and the proteins with similar functions, Factor H and C4bp (Chung,
LP et al., 1985, Biochem. J. 230: 133; Kristensen, T. et al., 19
86, J. Immunol. 136: 3407), and C3b and 4C, respectively.
Factor B, a component that forms an enzyme complex with b
And structural relationships of C2 with Ba and 2Cb fragments (Morley, BJ and Campbell, RD, 1984, EMBO J. 3:15
3; Mole, JE et al., 1984, J. Biol. Chem. 259: 3407; Bentley,
DR and Porter, RR, 1984, Proc. Natl. Acad. Sci. US
A.81: 1212; Gagnon, J., 1984, Philos.Trans.R.Soc.Lond.
B.Bio.Sci.306: 301) immediately after. However, SCRs have also been found in some non-complementary proteins (Campbell, RD and Bentl).
ey, DR, 1985, Immunol. Rev. 87:19; Lozier, J. et al., 1984, P.
roc.Natl.Acad.Sci.USA 81: 3640; Leytus, SP et al., 198.
6, Biochemistry 25: 4855; Kurosky, A. et al., 1980, Proc. Nat.
I. Acad. Sci. USA 77: 3388; Ichinose, A. et al., 1986, Bio-
(chemistry 25: 4633) (FIG. 7) shows that it does not necessarily represent a C3 / C4 binding structure.

Among proteins with SCRs, CR1 is unique in that it organizes this substrate structure and genetic units into higher-order structural units of the LHR. Analysis of a 14.5 kb BamHI fragment of genomic DNA associated with the expression of the S allotype indicates that at least one repetitive genomic unit in CR1 is a DNA fragment containing exons encoding at least 5 SCRs and their flanking introns. (Wong, WW et al., 1986, J. Exp. Med.
164: 1531). These studies also suggested that the S allele contained an additional copy of this genomic unit, as compared to the number present on the F allele. This observation was used in a trypsin peptide mapping study (Nickells, M.
W. et al., 1986, Mol. Immunol. 23: 661) and an LHR of -40-50.
Combined with the findings here, which correspond to a peptide of kD, we find that S allotype (290 kD)
The presence of additional LHR in F allotypes (250,0
(00 Dalton molecular weight)
Can be expected.

In addition to providing evidence for duplication phenomena, LHR
The sequence also suggests that a conversion event has occurred within the CR1 gene. LHR-B and -D are 67% identical to each other over their entire length, whereas LHR-C is LHR-C.
B and 99% identical in NH2-terminal 4 SCR, LHR-D and CO
91% identical in the OH-terminal 3 SCR. Such a mechanism would not have been achieved by a single recombination between identical parent alleles at the origin of the hybrid plasmid. Rather, the hybrid LHR may have resulted from genetic conversion such that sequences in the LHR-C precursor are replaced by sequences present in LHR-B or LHR-D (Atchison, M. and Adesnik, M., 1986, Pro
c.Natl.Acad.Sci.USA83: 2300). Also these LHR
The almost perfect identity and exact sequence of the homologous sequence in (FIG. 5) may have been maintained by mechanisms involving gene conversion. Analysis of the degree of homology between the intervening sequences of the CR1 gene segment encoding the LHR should determine whether the gene conversion strictly limited sequence divergence based on functional constraints.

Previous studies have suggested that CR1 is monovalent (Wilso
n, IG et al., 1982, N. Engl. J. Med. 307: 981)
Corresponds to a single C3b / C4b binding domain, thereby rendering the receptor multivalent and adapted to bind complexes bearing multiple C3b and C4b molecules. Alternatively, different LHRs correspond to binding of C3b and C4b, respectively (see Section 9 below), providing a structural basis for the combination of Factor H and C4bp activity in CR1. Finally, CR
1 of LHR are likely corresponding to structural domain useful in extending the CR1 as suggested from the plasma membrane by the submitted structural model (Figure 8), NH ₂ - terminal region SCR Is C3b as found for Factor H
And C4b (Sim, RB and DiScipio, RG, 1
982, Biochem. J. 205: 285; Alsenz, J. et al., 1984, Biochem. J.
224: 389). Activation of protein kinase C by phorbol esters induces CR1 phosphorylation in neutrophils, monocytes, and eosinophils (Changelian, PS and Fear
on, DT, 1986, J. Exp. Med. 163: 101), the 43 amino acid CR1 cytoplasmic domain has a sequence homologous to the site phosphorylated by protein kinase C in epidermal growth factor (Hunger, T. et al. Et al., 1984, Nature 311: 480; Davis,
RJ and Czech, MP, 1985, Proc. Natl. Acad. Sci. US
A.82: 1974). However, although this cytoplasmic sequence was found in three independent clones of the tonsillar library, it was most likely that of the B cell CR1, which was activated after activation by protein kinase C. Not phosphorylated (Changelian, PS and Fearon,
DT, 1986, J. Exp. Med. 163: 101).

7. Example: CR1 5 'cDNA sequence contains a fourth long homologous repeat By analysis of the partial cDNA sequence of CR1, three LHR, LHR-B, LHR-C, LHR- D, each comprising seven short consensus repeats (SCRs) of 65 amino acids characteristic of a C3 / C4 binding protein (see FIG. 6 above). In the examples described herein, we determined the fourth amino terminal LHR, LHR-A (Klickstei
n, LB et al., 1987, Complement 4: 180). Analysis of LHR-A revealed that LHR-A was 99% homologous to LHR-B in the five 3 'SCRs, but only 61% homologous in the two 5' SCRs. Was.

7.1. Materials and Methods 7.1.1. Construction of cDNA Library Description (CHirgwin, JM et al., 1979, Biochemistry 18: 5290;
Aviv, H. and Leder, P., 1972, Proc. Natl. Acad. Sci. US
A. 69: 1408; Ausubel, FM et al., 1987, Current Protocols i.
n Molecular Biolgy, John Wiley & Sons, New York) with the following modifications to selectively use primer-based cDNA libraries, λHH, from 3 μg of poly (A) ⁺ RNA purified from DMSO-induced cells. It was constructed. LK35.1, 35−
Mer oligonucleotide, 5'-TGAAGTCATC ACAGGATTT
C ACTTCACATG TGGGG-3 'was used instead of oligo (dT) _12-18 , and 40 μCi of α ³² P-dCTP was added during the second strand synthesis. One third of the cDNA was cloned into λgtll and a cDNA library was constructed from human tonsil poly (A) ⁺ RNA according to section 6.1.2 above. 750,000 mutually independent recombinants were obtained.

7.1.2. Clone Isolation, Probes, and DNA Sequence Analysis The probe used to screen the cDNA library was CR-1 (Wong, WW et al., 1985, Proc. Nat.
l.Acad.Sci.USA82: 7711) (ATCC No. 57331), CR-
2 (Wong.WW et al., Supra), CR1-4 (Wong, WW et al., 198
6, J. Exp. Med. 164: 1531), and nucleotides in FIG.
0.5 kb EcoRI of cDNA clone λH3.1 corresponding to 101-352
C, which is a 252 bp Sau3A fragment from the fragment
R1-18. Under high stringency conditions, CR1-18 is
Hybridizes only to cDNA encoding either the NH2-terminal SCR or the signal peptide. cDNA
The clone insert was inserted after subcloning the fragment into M13mp18 and M13mp19 (Yanisch-Perro
n, C. et al., 1985, Gene 28: 351), dideoxynucleotide method (Sanger, F. et al., 1977, Proc. Natl. Acad. Sci. USA 72:
5463).

7.2. Results A λgtll cDNA library using specific primers containing 7.5 × 10 ⁵ recombinants was prepared using cDNA synthesized from poly (A) ⁺ RNA from DMSO-derived HL-60 cells. did. These cells express only the F allotype of CR1 (Lublin, DM et al., 1986, J. Biol. Chem. 261: 573).
6), these are expected to have 4 LHRs (Lapat
a, MA et al., 1984, Nucl. Acids. Res. 12: 5707). The primer, LK35.1, was an antisense 35-mer corresponding to nucleotides 896-930 of the partial cDNA sequence of CR1 shown in FIG. This oligonucleotide is LHR under reverse transcription conditions.
-B, LHR-C and LHR-D were shown to hybridize. CR1 cDNA probe, CR1-1 and CR1-4
250 positive colonies were identified in a plate of 3.8 × 10 ⁶ unamplified recombinant phage screened with the mixture of 38 positive clones were picked and plaque purified. EcoRI from these clones
Southern blots of the digested DNA were analyzed for the 23-mer oligonucleotide, KS23.1, 5'-CTGAGCGTAC CCAAAGGGAC AAG-
Screened using 3 '. This oligonucleotide corresponds to nucleotide 76 of the partial CR1 cDNA sequence of FIG.
This corresponds to 3-785. The probe hybridizes at one site in the sequence encoding LHR-B under highly stringent conditions, but not in the sequence encoding LHR-C or LHR-D. Clone λH7.4
The inserts in FIG. 9 (1.0 kb, 0.9 kb and 0.4 kb)
EcoRI fragment and two of the larger, KS23.1-hybridizing fragments, indicating that this clone was 3 '5/7 LHR-A
And all LHR-B coding sequences. This finding confirmed that LHR-A is very homologous to LHR-B. Clone λH3.1 (No. 9
The figure) contained an EcoRI fragment with KS23. 1 in units of 1.0 kb and a 0.5 kb fragment that hybridized weakly with CR1-4 under high stringency. This clone is LHR
It is believed that -A is completed and contains additional 5 'sequences including SCR-1 and -2 and 0.1 kb of upstream sequence. Of the remaining 36 clones, all of them were CR1-1
And hybridized, but was not detected by the probe, CR1-18. This probe is LHR-B, -C, or-
252 bp Sau3A from the 0.5 kb EcoRI fragment of clone λH3.1 which encodes D and does not hybridize to the sequence
I fragment.

DNA sequence analysis of λH3.1 revealed that the oven reading frame continued to the 5 ′ end of the cDNA, indicating that the clone did not reach the translation start site. Therefore, the cDNA libraries, λHH and λS2T, were screened again using probe CR1-18, and λH10.3 and λT109.1 were identified, one clone from each. The EcoRI fragment of these clones that hybridized with CR1-18 was sequenced similarly to the inserts from those clones, λH3.1 and λH7.1. The synthesized sequence is shown in FIG. 1, where nucleotide number 1531 in FIG. 1 is nucleotide # 1 in FIG. HL
The overlapping sequences of the cDNA clones from -60 and the tonsil library are identical.

Immediately upstream of LHR-A, clones λT10,3 and λH10
9.1 is the same putative hydrophobic leader sequence (Vin Heijne,
G., 1986, Nucl. Acids Res. 14: 4683), which encodes 41 amino acids and includes an ATG that matches the consensus sequence NNA / GNNATGG submitted for the eukaryotic translation initiation site ( (Figure 10) (Kozak, M., 1986, Cell44: 283). A second ATG, located 6 codons upstream of the chosen ATG and just downstream of the in-frame stop codon, is less relevant to this consensus sequence. The first three amino acids of the leader sequence, such as CR1, MGA are the same as reported for CR2. The sequence of these two clones splits upstream of the ATG, and that from the λ10.3 clone is believed to correspond to a portion of the intervening sequence, as described for the other CR1 cDNA clones in Section 6 above.

Signal peptide cleavage is expected to occur between glycine-46 and glutamine-47 (Von Heij) ne, G., et al.
1986, Nucl, Acids Res.14: 4683 ), this is NH ₂ which is a block of CR-1 - end (Wong, WW, et al., 1985, J.Immun
ol.Methods 82: 303; Holeis, VM et al., 1986, Complement 3: 6.
3) suggests that pyrrolidone amide may be present. NH ₂ contained in these clones
The first two SCRs of the terminal LHR-A are only 61% identical to the corresponding region of LHR-B, while the SCR of LHR-A
3-7 is 99% identical to the corresponding SCR of LHR-B (No. 10
Figure). Comparison of LHR-A with LHR-C reveals that only each of the third and fourth SCRs are highly homologous (99% identical). LHR-A and -D totaled only 68
% Identity, but between the 6th SCR of each LHR has a maximum of 81% identity. Therefore, the 5'cDN of CR1
The completion of the A sequence results in the F allotype comprising 2039 amino acids, a signal peptide of 41 amino acids, each of 7
Includes four LHRs consisting of SCRs, two COOH-terminal SCRs, a 25 residue transmembrane region and a 43 amino acid cytoplasmic domain. There are 25 potential N-linked glycosylation sites.

7.3. Discussion The primary structure of the NH2-terminal and signal peptide of the F allotype of CR1 was deduced by isolation and sequencing of a 5 'cDNA clone. The highly repetitive nature of the CR1 sequence made it essential to develop suitable methods for the preparation and identification of cDNA clones encoding this region of the receptor. It is known to hybridize with LHR-B, -C and -D under the conditions of reverse transcription.
CDNA using mer oligonucleotides as primers
A library was prepared. This primer, LHR-B
It was considered possible to hybridize with LHR-A which was predicted to be very homologous to LHR-A (see section 6 above).
A suitable cDNA clone was identified using another oligonucleotide, KS23.1.
And only hybridizes with it, thereby increasing the probability of finding a 5 'cDNA clone. Almost all CR1
Two clones were found that encompassed the sequence of residues of which one, the Sau3A I fragment, CR1-18, had a sequence sufficiently unique to use it to identify the remaining five clones (No. 9). , 10).

The 250 bp probe from the 5 'region of LHR-A hybridized not only to the 7.9 and 9.2 kb CR1 transcripts, but also to the 2 kb transcript in human tonsil RNA under stringent conditions. Such cross-hybridized mRNA was observed with CR1 cDNA probes from other LHRs or in Northern plots of RNA from dimethyl sulfoxide derivative LH-60 cells and HSB-2 T lymphoblast cells. Did not. Thus, CR1 contains sequences homologous to two additional B cell proteins, one encoded by this newly recognized mRNA and the other CR2.

8. Example: Expression of Recombinant Human CR1 As described above, a 7.0 kb human CR1 cDNA clone was isolated, which has an open reading frame encoding 2039 amino acids (FIG. 1). The precursor form of the proposed receptor consists of a signal peptide of 41 amino acids, four long homologous repeats (LHRs) of 450 amino acids (each LHR comprises seven short consensus repeats (SCRs)), 65 amino acids The two COOH-terminal SCRs, a 25 amino acid transmembrane domain, and a 43 amino acid cytoplasmic region. Thus, the CR1 F allotype contains 30 SCRs. NH ₂ -terminal L
HR, LHR-A (see section 7 above) is 61% identical to the corresponding region of LHR-B in the first two SCRs and 99% identical in the 5 SCR at the COOH-terminal. 8 CR1s
Splice restriction fragment of cDNA clone to 6.9k
A full-length construct of b was made, placed downstream of the mouse metallothionein promoter or cytomegalovirus promoter and transfected into L (mouse) or COS (monkey) cells. Recombinant cell surface CR1 was detected by indirect radioinomic assay and immunofluorescence. Parent vector (CR1 ^-) in cells transfected with only the antigen was detected. Immunoprecipitation of transfected, surface- ^125I- labeled COS (monkey) cells with anti-CR1 monoclonal antibody and analysis by non-reduced sodium dodecyl sulfate (SDS) polyacrylaminogel electrophoresis were performed using human erythrocyte-derived F allotype. A band with a molecular weight of 190,000 daltons migrated with it. Expression of the correct molecular weight recombinant CR1 antigen (K
lickstein, LB et al., 1988, FASEB J.2: A18 33)
Demonstrates that the DNA contains the complete coding sequence for human CR1.

8.1. Plasmid pBSABCD containing the complete CR1 coding sequence
Here, the construction of the vector encoding the full-length (SCR1-30) CR1 protein and the plasmid pBSABCO will be described.

2.3 kb insert from cDNA clone λT8.2 (Klic-kstei
n, LB et al., 1987, J. Exp. Med. 165: 1095; see section 6 above)
Was subcloned into pUC18 as an EcoRI fragment so that the 5 'end was
Close to Hind III. This plasmid was named p188.2. p188.2 was cut with Apa I and Hind III and the large 4.7 kb fragment containing the CR1 sequence from SCR26 to the 3 'untranslated region and the vector sequence was gel purified.

Insert from cDNA λT50.1 (klickstein, LB et al., 198
7, J. Exp. Med. 165: 1095, see section 6 above) was subcloned into M13mp18 as an EcoRI fragment. This phage was called 18R50.1. DNA from the cloned form of this clone was cut with Apa I and Hind III, and CR1 SCR1
The 1.45 kb fragment containing 8-25 was isolated and p188.2
Ligated to a 4.7 kb fragment from E. coli D
H5α was transformed. This plasmid was designated as p8,250.1.

0.75 kb and 0.93 kb EcoR from cDNA clone λT8.3
I fragment (Wong, WW et al., 1985, Proc. Natl. Acad. Sc
iUSA22: 7711) was subcloned into plasmid pBR327. These subclones were called pCR1-1 and pCR1-2, respectively, which contained SCR11-14 and SCR17-21, respectively. The 0.75 kb pCR1-1 fragment purified from each of the EcoRI inserts was digested with SmaI,
The digest was ligated to pUC18 DNA cut with EcoR I and Sma I. One of the subclones with a 0.5 kb insert corresponding to SCR12-14, p181-1.1, was isolated.

The 0.93 kb fragment of pCR1-2 was digested with HindIII,
Binds to pUC19 cut with EcoR I and Hind III and binds to SCR17
One of the subclones with a 0.27 kb insert, p191-2.1, was isolated.

cDNA clone λ6.1 (see Section 6 above; Klickstein, L .;
B. et al., 1987, J. Exp.Med. 165: 1095; Wong, WW et al., 1987, J.
Exp.Med. 164: 1531) was digested with EcoRI and the 0.37 kb fragment corresponding to SCR15 and 16 of CR1 was subcloned into pBR322. This clone was named pCR1-4. Clone p181-1.1 was cut with EcoR I and Sca I, 1.4 kb
The fragment was isolated. Clone p191-2.1 (Klickst
ein, LB et al., 1987, J. Exp. Med. 165: 1095; see section 6 above), digested with EcoRI and ScaI, isolated the 2.0 kb fragment and ligated to the 1.4 kb fragment from p181-1.1. The mixture was transformed into E. coli DH5α. The resulting plasmid was named p1-11-2. Plasmid p1-11-2 was digested with EcoR I and digested with 0.
The 37 kb insert fragment was inserted by ligation. The resulting plasmid was used for transformation of E. coli DH5α.

A subclone containing the 0.39 kb BamHI-HindIII fragment was selected. This plasmid is called p142,
It contained CR1 SCR12-17. 3.5kb from p8.250.1
The EcoR I-Hind III insert fragment was transferred to pGEM3b. This plasmid was designated as pG8.250.1. The 1.2 Hind III fragment from p142 was purified and ligated to pG8.250.1 which had been pre-cut with Hind III. A subclone containing the 2.4 kb PstI-ApaI insert was selected, and therefore the correct orientation was selected. This plasmid was called pCD and contained the CR1 sequence from SCR12 to the 3 'end. cDNA clone λ5'7.1 (Klickstein, LB et al., 1987, Complemen
t4: 180; see section 7 above), cut with PstI, and isolate a 1.35 kb fragment corresponding to SCR6-12, PstI-cut pCD
Bound. The mixture was transformed to 1.35 kb and 1.
A subclone containing the 1 kb Hind III fragment was selected.

cDNA clone λ5'3.1 (Klickstein, LB et al., 1987, C
omplement4: 180; see section 7 above) with EcoRI.
The digest was ligated to EcoRI-cut pUC18. A subclone, p3.11-1, was isolated, which corresponds to SCR3-7
Although containing a 1.0 kb insert, this insert was gel purified. cDNA clone λ5′10.3 (Klickstein, LB et al., 198
7, Complement 4: 180; see section 7, supra), was cut with EcoRI and the 0.63 kb insert containing SCR1 and 2 was subcloned into pUC18. This clone was named p10.3.5. Plasmid p10.3.5 was partially digested with EcoRI and the 3.4 kb fragment corresponding to the linear plasmid was isolated, p3.11
-1 from the 1 kb fragment. Subclone p
LA was picked up and contained a 1.3 kb Pst I fragment at the correct site for insertion and orientation.

cDNA clone λT109.4 (Klickstein, LB et al., 1987, Co.
mplement4: 180; see section 7 above), digest with EcoR I and give pU
It was subcloned into C18. Leader sequence and SCR1
0.55kb Eco corresponding to 5 'untranslated region through and 2
A subclone containing the RI fragment was selected.
Plasmid p109.4 was cut with Pst I and PspM II and a 3.0 kb fragment containing the vector, leader sequence, and SCR1 was isolated. The fragment was ligated to a 0.81 kb PstI-BspMII fragment from pLA containing SCR2-5. This new plasmid was named pNLA.
Plasmid pNLA was partially digested with EcoR I and completely digested with Pst I, and the 1.1 kb EcoR I-Pst I fragment containing the CR1 sequence leading to SCR5 was isolated from the leader sequence, and pBlue
script KS + (Stratagene San Diego, CA)
An Xho I site was placed at the 5 'site of NA. This plasmid is called p
Called XLA.

Plasmid pBCD was digested with EcoR V and then partially digested with Pst I to isolate a 6.0 kb Pst I-EcoR V fragment containing the CR1 sequence from SCR6 to the 3 'untranslated region.
Bound to I + SmaI-digested pXLA. The resulting bacterial expression plasmid contained the complete CR1 cDNA coding sequence and was named pBSABCD.

8.2. Construction and Assay of the Mammalian Expression Vector Plasmid piABCD Containing the Complete CR1 Coding Sequence Plasmid pBSABCD was digested with XhoI and NotI, and the insert was inserted into expression vector CDM8 (previously cut with these restriction enzymes). Seed, B., 1987, Nature 329: 840-842)
Was ligated downstream from the CMV promoter at a 4.4 kb fragment. The resulting construct was named piABCD (No. 11
Figure). Alternatively, a 6.9 kb XhoI-NotI fragment was ligated downstream from the metallothionein promoter in an expression vector, pMT.neol, also previously cut with these restriction enzymes. The resulting construct was named pMTABCD (FIG. 11).

Sheep erythrocytes (EA) sensitized with rabbit antibodies, and limited amounts of C4b [EAC4b (lim)], and 12,4 cells per cell
000 cpm of ¹²⁵ I-C3b [EAcC4b (lim), 3b] was added to EAC4b (li
m) (Diamedix) was prepared by sequential treatment with C1, C2 and 125I-C3, followed by incubation at 37 ° C for 60 minutes in gelatin veronal-buffered saline containing 40 mM EDTA. Alternatively, methylamine-treated C3
[C3ma)] was covalently bound to sheep E (erythrocytes) treated with 3- (2-pyridyldithio) propionic acid N-hydroxysuccinimide ester (Sigma) (Lambri)
s, JD et al., 1983, J. Immunol. Methods 65: 277). EAC4b was prepared using purified C4 (Hammer, CH et al., 1981, J. Bio.
l. Chem. 256: 3995).

Both piABCD and pMTABCD are COS (monkey) by the DEAE-diethylaminoethyl) -dextran method.
Cells were transfected. By immunofluorescence using an anti-CR1 monoclonal antibody, YZ-1; and immunoprecipitation of ¹²⁵ I-labeled cells followed by non-reduced SD
By S-PAGE; and by rosetting on sheep erythrocytes coated with C3b (Fearon, DT, 1980, J. Ex.
p.Med. 152: 20), recombinant CR1 was detected on the surface of transfected cells. This SDS-PAGE was immunoprecipitated from donor human erythrocytes homozygous for the F allotype.
A protein having the same mobility as CR1 was shown (Wong.
WW et al., 1983, J. Clin Invest. 72: 685). The fact that the electrophoretic mobilities of the native and recombinant CR1 were the same confirmed that the CR1 F allotype contained SCR1-30.

Furthermore, mouse L cells were subjected to the DEAE-dextran method (Ausu
bol, FM et al., 1987, Current Protocols in Molecular Bi.
ology, Seidman, JG and Struhl.K. Ed., John Wiley &
Sons, New York; Seed, B., 1987, Nature 329: 840), in duplicate, 0.2 or 4 μg of piABCD or pMTABC.
D and 2 μg of pXGH5 (Selden,
RF et al., 1986, Mol. Cell. Biol. 6: 3173). Two days later, cells are collected and YZ1
CR1 expression was assayed by binding of a monoclonal anti-CR1 antibody. There was a dose-response relationship between recombinant plasmid DNA and expression of the CR1 antigen (Table II).

Plasmid piABCD indicated about three times more expression of the CR1 antigen than pMTABCD. Except for plate 5, the growth hormone concentration in the medium varied within a range of less than twice. Additional experiments show that piABCD has three times more CR1 in COS cells than L cells.
Has been shown to direct transient expression of the antigen.

When assessed by indirect immunofluorescence of cells stained with ZY1 anti-CR1 mAB, CR1 was transfected COS
It was present in the cell surface closter (Figure 12). Such an analysis of recombinant CR1 on COS cells mimics the distribution of wild-type CR1 on human leukocytes (Fearon et al., 1981, J. Exp. Me.
d.153: 1615).

The molecular weight of recombinant CR1 was determined by surface iodination of COS cells transfected with piABCD, immunoprecipitation reaction of cell lysate with Sepharose-YZ1, SDS-PAGE and autoradiography. Recombinant CR1 has a non-reducing molecular weight of 190,000, which is equal to the F allotype and smaller than the S allotype of erythrocyte CR1 (No. 14).
Figure).

The C3b and C4b binding functions of recombinant CR1 were assayed by rosetting of transfected COS cells with EAC4b or EAC4b (lim), 3b. In 31 separate transfections, 5% -50% of COS cells transfected with plasmid piABCD contained 5%
Bound to one or more EAC4b or EAC4b (lim), 3b (FIG. 13). COS cells expressing CR1 are EAC4b (li
m), did not rosette with 3bi (although this intermediate formed rosettes with CR2-expressing RajiB lymphoblast cells).

8.3. Expression of CR1 Fragment An expression vector (deletion mutant) encoding a part of the CR1 coding sequence was constructed as follows, and it was confirmed that each of them expresses a CR1 insert when transformed into COS cells. I found it. The CR1 fragment was expressed as a cell surface protein.

8.3.1. Construction of deletion mutants piBCD, piABD, piACD, piAD, piBD, piCD and piD The construction of such a mutant involves the long homology of four CR1 repeats near each amino terminus (LHR). Contains a single Bsm I site, CR1 cDNA and Bluescript vector (Stra
tagene, San Diego, CA).

Ten micrograms of plasmid pBSABCD was partially digested with 50 units of the restriction enzyme BsmI for 45 minutes, and the digest was fractionated by agarose gel electrophoresis. 8.55 kb, 7.20 kb and 5.85 kb DNAs corresponding to the linear segments of the parent plasmid with one, two or three LHRs deleted, respectively
The plug was purified. Each of the three fragments was ligated to itself, and the ligation was used separately to transform competent E. coli DH5α to ampicillin resistance.

The 8.55 kb fragment binds pBSABCD to two adjacent Bsm I
Resulted from cutting at the site. Thus, after ligation, there may be three product plasmids, pBCD, pACD or pABD. Here, the capital letters remained on the plasmid
Indicates LHR. These are distinguished by restriction mapping using SmaI. DNA was prepared from 12 colonies, digested with Sma I, and separated by agarose gel electrophoresis. Five colonies have two SmaI fragments of 2.5 kb and 6.1 kb, corresponding to a deletion of the LHR-A coding sequence, and
Hit BCD. Three colonies have a single linear fragment of 8.5 kb, corresponding to pACD. 4 colonies 1.2kb and 7.4
with two Sma I fragments of LHR-C
A deletion of the coding sequence is expected, indicating pABC. The 5.6 kb insert of each of these three constructs was double digested with XhoI and NotI, then gel-purified, and ligated to the expression vector CDM8 that had been previously digested with the same restriction enzymes and then gel-purified. E. coli
DK1 / P3 was transformed with the ligation mixture, and DNA was prepared from each of the 5 colonies. CR1 deleted in expression vector
The presence of the cDNA insert was determined in each case by Sac
As shown by I digestion, it showed the two expected fragments of 4.20 kb and 5.75 kb. These plasmids were named piBCD, piACD and piABD.

The 7.20 kb fragment obtained from the partial digestion of pBSABCD has Bsm I at three flanking sites or, for similarly large fragments, two sites with one uncleaved site between the two sites. It was the result of the digestion, and thus two products, pAD and pCD, could be obtained after transformation. These can be distinguished by double digestion with Xho I and Pst I, which in the case of pAD yields two fragments of 1.0 kb and 6.2 kb, and for pCD yields a linear fragment of 7.2 kb.
The 4.2 kb insert from each of these plasmids was
After double digestion with t I, gel-purified and C
It was subcloned into DM8. The presence of the deleted CR1 cDNA in the expression vector was shown by double digestion with PstI and BglII. Clone piAD is 2.
PiCD had a single fragment of 8.6 kb, with 4 kb and 6.2 kb fragments.

The 5.85 kb fragment obtained from the BsmI digestion of pBSABCD corresponds to the complete digest, and one clone pD is E. coli DH.
Obtained after 5α transformation. This is Hind III and Bgl
Confirmed and expected by double digestion with II
3.7 kb and 2.2 kb fragments were generated. After double digestion of the 2.9 kb insert of the clone with Xho I and Not I,
Gel-purified and ligated to expression vector as described above.
Hind III digestion of the resulting piD clone yielded the expected 7.3 kb fragment, and double digestion with Xho I and Bgl II gave 2.2 kb and 5.1 kb fragments.
The Sac I digest gave the expected 1.5 kb and 5.8 kb fragments.

Plasmid pBD was prepared by BsmI partial digestion of pBCD. Linear shape corresponding to cleavage at two adjacent Bsm I sites
The 7.2 kb fragment was gel purified, ligated to itself as described above, and transformed E. coli DH5α to ampicillin resistance. pBD was identified by the presence of the 1.2 kb and 6.0 kb fragments in the SmaI digestion. Insert the 4.2kb insert into Xho
After double digestion with I and Not I, purify and
Transferred to DM8. Clone piBD was confirmed by observation of the expected 0.8 kb and 7.8 kb fragments after Hind III digestion.

COS cells transiently expressing the piABCD, piBCD, piCD, and piD constructs, respectively, were surface labeled with ¹²⁵ I and immunoprecipitated with anti-CR1 antibodies. On SDS-PAGE after reduction, piA
The products of the BCD construct migrated similarly to the F allotype of CR1, while the deletion mutants showed a gradual decrease of about 45,000 daltons indicating deletions of 1, 2, and 3 LHR, respectively (17th. Figure).

8.3.2. Deletion mutant piP1, piE1, piE2, piE-2, piU1, piU-2
Plasmid piABCD was completely digested with BstE II and 1.35 kb
(Tablet) and the two fragments of 8.6 kb were gel purified, mixed, ligated and transformed into E. coli DK1 / P3 to ampicillin and tetracycline resistance.
Colonies were screened by hybridization with CR1 cDNA probe CR1-4 (see section 8.1 above).
Strongly positive clones were picked and further screened by digestion with SmaI. piE1 is 2.7kb and 7.3kb
Were identified by the presence of two fragments. piE2
Was identified by a single 10.0 kb linear fragment.
piE-2 was identified as a weak CR1-4 positive clone containing a single 8.6 kb SmaI fragment.

Plasmid piP1 was obtained by complete Pst I digestion of piABCD and gel purification of a large 10.0 kb fragment. This fragment was ligated and E. coli DK1 / P3 was transformed with the mixture. The resulting plasmid, piP1, contained a single 10.0 kb SmaI fragment.

Plasmids piU1 and piU-2 were prepared by initial transformation of dcm ⁻ strain GM271 / P3 with plasmid pXLA. This DNA was double digested with Stu I and Nco I,
The 3.3 kb fragment was gel purified. Plasmid pBSABCD
Was partially digested with NsiI and the resulting 4 bp 3 ′
The overlap was removed by treatment with the Klenow fragment of E. coli DNA polymerase I. The DNA was then completely digested with Not I and the 5.4 kb and 4.0 kb fragments gel purified. These are 3.3 kb from pXLA
The ligation mixture was ligated to the Stu I-Not I fragment and E. coli DH5α was lightly transformed to ampicillin resistance. Screening of colonies by hybridization with CR1 cDNA probe CR1-4, positive clones
Checked further by Hind III restriction digestion, which produced three fragments of 0.8 kb, 1.3 kb and 6.5 kb for pU1 and two fragments of 0.8 kb and 6.5 kb for pU-2. Nsi branded with Stu I
The I splice was confirmed to be in reading frame by DNA sequencing of these plasmids. 5.6k each
The b and 4.2 kb pU1 and pU-2 inserts were gel purified after XhoI- and NotI double digestion and ligated to the expression vector CDM8 as described above. Clones, piU1 and piU
The structure of -2 was confirmed by restriction digestion with Xho I and Pst I, with two expected 1.2 kb and 8.8 kb fragments for piU1 and a linear 8.7 kb fragment for piU-2. occured.

Plasmid piA / D was prepared by first complete digestion of piABCD with PstI. The Pst I digest was then partially digested with Apa I and the 3 'overlap was removed with the Klenow fragment of E. coli DNA polymerase I. The DNA was then fractionated by agarose gel electrophoresis, and the 7.5 kb fragment was isolated, ligated and used to transform E. coli DK1 / P3 to ampicillin and tetracycline resistance. The construction was confirmed by double digestion with KpnI + SacI and the expected 0.8 kb, 1.5 kb, 1.7 kb and 3.3 kb
Yielded 4 fragments.

9.Examples: Identification of C3b and C4b Binding Domains9.1.Assays and Results Plasmids piABCD, piAD, piCD and piD containing the LHR indicated by the acronym of their name were transformed into COS cells and Used in assays to evaluate the ability of the encoded CR1 fragment to bind C3b or C4b. The binding assay uses full-length CR on their cell surface.
Express one molecule or CR1 deletion recombinant (transient expression)
This was done by observing erythroid rosette formation resulting from the binding of C3b or C4b coated erythrocytes by COS cells.
Transfection cells 1-4 × 10 ⁶ / ml into C3 or C4
20 ° C. for 60 minutes in 0.02 ml with 2-6 × 10 ⁸ / ml of loaded erythrocytes
Incubated. The percentage of transfected cell forming rosettes was evaluated microscopically by transfected cells recorded as one rosette with at least 5 adherent red blood cells. The results are shown in Table III.

In each of three separate experiments, the proportion of COS cells expressing the full-length piABCD construct rosettes with EC3 (ma) was determined by immunoassay using either YZ1 monoclonal anti-CR1 antiserum or rabbit anti-CRR antiserum. Similar to the fraction with detectable recombinant receptor as assessed by fluorescens (Table III). In contrast, cells expressing piD did not form rosettes. This indicates that the C3-binding site must be or is required in LHR-A, -B or -C. One site is piBD or
It was shown to be present in both LHR-B and -C by demonstrating that cells expressing either of the piCDs form rosettes with EC3 (ma). Cells expressing piAD, piA / D or piE-2 did not have comparable C3-binding function. The piE-2 construct is the first two SCs of LHR-C.
Only at the point having SCR-1 and -2 of LHA instead of R
Since differs from the ICD, the function of C3- binding site in LHR-C These NH ₂ - would require a terminal SCR.

The percentage of COS cells expressing the full-length piABCD recombinant that rosettes on EC4 (ma) was less than the fraction rosettes on EC3 (ma). This is probably
Less C4 (ma) per erythrocyte (Part III
Tables) or reflect fewer C4-binding sites per receptor. Deletion recombinants piAD, piA / D and piE-2 constructs having all or part of LHR-A are
It bound EC4 (ma) better than the deleted recombinants piBD and piCD; piD lacked this function. Therefore, CR1 C
The 4-conjugate may be present initially in LHR-A, although secondary sites may be present in LHR-B and -C. piE-2
The better rosetting ability of the construct compared to that of piCD suggests that SCR-1 and -2 of LHR-A are related to the C4-binding site.

Radioimmunoassay of YZ1 monoclonal anti-CR1 antibody binding showed significant uptake by COS cells expressing piABCD, piAD, piBD and piCD constructs (IV
table). 5 NH ₂ -terminal SCRs of LHR-A and 3 of LHR-D
Cells transfected with piD or piA / D consisting of two COOH-terminal SCRs, the product of these constructs bound polyclonal anti-CR1 antiserum, but YZ1
It did not bind to anti-CR1 antibody (Table IV). Therefore, Y
The Z1 epitope is repeated in LHR-A, -B and -C,
Of HR-A NH ₂ - not present in terminal SCR, and LHR-D
Is absent or unavailable in

9.2. Discussion A conserved BsmI site found in the middle of the coding sequence of the first SCR in each LHR closely corresponds to the border of the LHS and is required for open reading frame and putative disulfide bond formation It was possible to construct a series of deletion recombinants retaining the proper positions of the four cysteines (FIG. 16). The C3 (ma)-and
Comparison of C4 (ma) -binding functions distinguished not only LHRs with these specificities, but also the SCRs that are critical for determining ligand specificity. Therefore, piAD, piA / D and piE-2 forms, but not piD forms, transfected COS cells with EC4 (m
ability to mediate rosette formation between a) the two NH ₂ in LHR-A - showed that the terminal SCR contains sites for interaction with this complement protein (Table III) . This site was established because transfectants expressing piAD and piA / D were able to bind EC3 (ma) (III
Table), only relatively specific for C4 (ma). Demonstrated by the factor I-cofactor function for rosette assay and cleavage of piBD and piCD and C3 (ma) (Table III; FIG. 18) piBD and piC
The C3 (ma) -binding function of the receptor encoded by the D construct indicates that C3 (ma) is present in the first two SCRs of these LHRs.
It was shown that there was a site specific to These sites were also able to interact with C4 (ma) (Table III). Thus, LHR-A, -B and -C have selective but overlapping C4- and C3-binding activity.

Alternatively, COS cells expressing piBD and piCD are
Ability to bind to C4 (ma) is, NH ₂ - in which nucleotides encoding the terminal 36 amino acids were generated by transferring from SCR-1 of LHR-A by the coupling of the Bsm I fragment LHR-B and -C There will be. However, these 36 amino acids alone did not confer a C4-rosette forming function on the piD product. We cannot rule out the secondary function of LHR-D in these reactions.
This LHR-D was present in all of the constructs assayed for function. The discovery that there are three distinct ligand recognition sites in CR1, two for C3b and one for C4b, despite the fact that each receptor molecule has a relatively low affinity for monovalent ligands (Arnaout 1983, Immunology 48: 229) show that complexes carrying multiple C4b and C3b molecules could be effectively bound. This finding also provides an explanation for the inability of soluble C4b to inhibit rosette formation between C3b-bearing erythrocytes and human B lymphoblastoid cell lines (Gaither, TA et al., 1983, J. Am. Immunol. 131: 399).
Possible ligands to which CR1 would be particularly suited would be the molecular complexes C4b / C3b and C3b / C3b formed upon activation of the major and alternative pathways, respectively. Since there are distinct binding sites in three of the four LHRs, CR1 structural allotypes that differ by their number of LHRs will have significant functional differences caused by changes in the number of ligand binding sites. In vitro studies were performed on F, S and F '(A, B
And C) Although not reporting different binding activities of allotypes, perhaps small F 'allotypes with only 3 LHRs may have impaired ability to elucidate immune complexes. The F 'allotype has been reported to be probably associated with systemic lupus erythematosus (van Dy
ne, S. et al., 1987, Clin. Exp. Immunol. 68: 570).

10. Examples: Display of Factor I Cofactor Activity Recombinant CR1 protein and specific fragments thereof both on the cell surface and in solubilized form have been shown to have C3b Factor I cofactor activity.

Assays for Factor I cofactor activity are described in published procedures (Fearon, DT, 1979, Proc. Natl. Acad. Sci. U.
SA76: 5867).

To assay the Factor I cofactor of solubilized CR1 and fragments, cell surface CR1 protein and fragments were solubilized with Nonidet P-40 and the lysate was coupled with an anti-CR1 monoclonal antibody YZ-1 coupled to a Sepharose gel. Immunoprecipitated. Detergent lysate of 1 × 10 ⁶ transfected cells was added to Sepharose
Immunoprecipitation was performed sequentially with UPC10 anti-levan and Sepharose-YZ-1. The washed beads were then added to 0.05 ml PBS, 0.
Immunoprecipitates were assayed for Factor I cofactor activity by incubating with 0.5 μg of ¹²⁵ I-C3 (ma) in 5% NP-40 and 200 ng of Factor I at 37 ° C. for 60 minutes. After incubation, the supernatant containing the radiolabeled C3 (ma) was analyzed by SDS-polyacrylamide gel electrophoresis and autoradiography. Factor I cofactor activity was demonstrated by the appearance on the autoradiogram of a low molecular weight form of the alpha chain of C3 (ma) resulting from proteolytic cleavage by Factor I.

Factor I cofactor for cell surface CR1 and fragments (piABCD, piAC, piBD, piCD or pi
D) 0.5 μg of transfected COS cells carrying
Incubated with ¹²⁵ I-C3 (ma) and 0.2 μg Factor I (Fearon, DT, 1977, J. Immunol. 119: 124
8).

The factor I-cofactor activity of the cell surface recombinant CR1 is shown in FIG. Factor I cleaved the alpha chain of C3 (ma) into fragments with molecular weights of 76,000 and 46,000 only in the presence of immuno-solidified recombinant CR1 or Factor H (FIG. 15). The region corresponding to the band from the autoradiogram was excised from the gel and assayed for ¹²⁵ I to determine the amount of cleaved alpha chain. In the presence of Factor H, 91% of the alpha chain was cleaved, but in the presence of increasing amounts of recombinant CR1,
%, 41% and 55% were cleaved. COS cells were transfected with CDM8 vector alone containing some endogenous Factor I-cofactor activity, but this increased function was evident by COS transfected with piABCD, piBD and piCD (FIG. 18). ). piAD
Alternatively, enhanced cleavage of ¹²⁵ I-C3 (ma) was not observed in COS cells transfected with piD. Thus, of these constructs, only the deletion recombinants piBD and piCD, which conferred C3 binding ability on COS cells, also had Factor I-cofactor activity for C3 cleavage.

Table V shows the results of assaying Factor I-cofactor activity by cell surface and soluble forms of CR1 and its fragments.

As shown in Table V, the expression of piABCD (encoding full-length CR1 protein), piBD (encoding LHR-B and -D) or piCD (encoding LHR-C and -D) is a C3b factor I CR1 with cofactor activity
The product was produced. Thus, the data in Table V provides proof that the CR1 protein or a fragment thereof can promote complement inactivation.

11. Example: Expression of recombinant soluble CR1 CR1 cDNA was modified by recombinant DNA manipulation to
A soluble form of one fragment (sCR1) was generated. sCR1
The construct was expressed in a mammalian system where the expressed protein was secreted from the cells. Large amounts of soluble polypeptides were produced, which did not need to be solubilized to obtain as a solution, in contrast to the membrane-bound form of the CR1 protein.

11.1. Materials and Methods 11.1.1. Enzymatic digestion All restriction digests, linker ligation, and T4 DNA
Ligase and E. coli DNA polymerase reactions were performed according to the manufacturer's instructions (New England Biolabs, Inc., Beverley, Mass.). Morrison, D. E. coli DH1 or DH5α
A., 1979, Meth. Enzymol 68: 326-331. Competent bacterial cells were isolated from Maniatis, T .;
Et al., 1982, Molecular Cloning, ALaboratory Manual, Cold
Spring Harbor Laboratory, Cold Spring Harbor, New Y
The DNA was transformed with ork. Plasmids were purified by the alkaline lysis or boiling method (Maniatis, T. et al., Supra).

11.1.2. Isolation of DNA Fragment The DNA fragment was agarose (BioRad, Richmond,
(A) The gel was purified as follows. The right DNA
The band was excised from the gel with a razor, agarose slices were placed on paraffin pieces, sliced into very small sections and transferred to fresh paraffin pieces. The agarose pieces were crushed and the agarose was transferred to 1.5 ml tubes. The same volume of phenol (ultra pure, BRL, Gaithersburg, MD) was added, the mixture was spun, then frozen at -70 ° C for 10 minutes and centrifuged for 10 minutes. The aqueous phase was further extracted twice with phenol / chloroform (1: 1) and twice with chloroform.
The DNA is then ethanol precipitated, the pellet washed,
It was dried under vacuum and resuspended in 10 mM Tris-HCl, pH 7.0, 1 mM EDTA.

The DNA fragments are converted to low gelling temperature agarose (FMC, Ccr
p., Rockland, ME) as follows. The appropriate DNA band was excised from the agarose gel, placed in a 1.5 ml tube and thawed at 65 ° C. for 15 minutes. Liquefied gel is converted to 0.1% sodium dodecyl sulfate (SDS, ultrapure, BRL, Gaithersberg, MD)
Extracted with phenol containing. The aqueous phase was further extracted once with phenol-SDS and twice with chloroform. The DNA is then ethanol precipitated in 2.0 M NH ₄ in acetate,
Dried and resuspended in water.

11.1.3. Transfection into Mammalian Cells DNA was transfected into mammalian cells by CaPO ₄ precipitation and the glycerol shock method of Graham and var der Eb (1973, Virology 52: 456-467). DUX
B11 CHO cells were incubated with the DNA-calcium phosphate preparation for 4-6 hours before being subjected to glycerol shock by removal of the growth medium by aeration and addition of 5 ml of 2% glycerol DMEM medium for 1 minute. The cells are then washed twice with complete alpha-MEM and 48
Incubated for hours.

11.1.4. CHO transfectant cell culture DUX B11 CHO cell transfectants were isolated from 10% dialysed fetal calf serum (Gibco) without nucleosides and 4 mM
-Grow in DHFR (dihydrofolate reductase) selective medium consisting of alpha-MEM medium (Gibco) supplemented with glutamine. Increasing amounts of methotrexate (Sigma, # A-
Amplification was performed by growing cells at a concentration of 6770, Amethoprotein) (Kaufman, RJ et al., 1985, Molec. Cell Bi.
ol. 5: 1750-1759).

11.1.5. ELISA for measuring soluble CR1 levels 11.1.5.1. CR1 standard A detergent lysate of hemoglobin-free red blood cells (RBC) ghost was used as a CR1 standard in ELISA (enzyme-linked immunosorbent assay). Ghosts were prepared as previously described (Wong.WW and Fearon.D.
T., 1987, Meth. Enzymol 150: 579-585). Briefly, the Red Cross obtained expired whole blood. Red blood cells
After washing three times with PBS, 6 volumes of hypotonic lysis buffer (10m
M Tris pH 8, dissolved in 0.1 mM PMSF (phenylmethylsulfonyl fluoride), 0.1 mM TPCK (tosylamide-phenylethylchloromethylketone), aprotonin, 2 mM EDTA. Ghosts were washed several times with lysis buffer, counted in a hemocytometer, divided into small portions and frozen at -70 ° C until needed. For CR1 ELISA, ghost was dissolved in a solubilization buffer (10 mM Tris pH 8,50 mM KC1, 0.2% NPO ₄ O, 0.3% DOC, 6.2 mM).
PMSF, 0.2mM iodoacetamide, aprotonin, 0.1mM
Diluted to 1.6 × 10 ⁸ ghost / ml in TPCK, 2 mM EDTA, 0.2% NaN ₃ ) and serially diluted to 2.5 × 10 ⁶ ghost / ml for use as a standard in ELISA. Absorbance at 90 nm was plotted and all unknown sample manipulations were correlated with the plot to give ghost equivalents / ml.

1.1.5.2.CR1 ELISA Immunlon-II plate was incubated at a concentration of 0.4 μg / ml in PBS.
CR1 monoclonal antibody (clone J3D3, AMAC IOT 17)
100 μl / well (Cook, J. et al., 1985, Molec.
mmunol. 22: 531-538) and incubated at 4 ° C. overnight. Next, the antibody solution is discarded, and the blocking solution (1.0% B
The plate was blocked by adding 300 μ / well (in SA, PBS) and incubated at 37 ° C. for 2 hours. After blocking, use the plate as it is,
Or stored at 4 ° C. until needed.

Plates were washed 3 times with PBS containing 0.05% Tween-20.
Washed twice. Samples were added in duplicate at 100 μ / well and incubated at 37 ° C. for 2 hours. In some cases, samples were diluted with solubilization buffer. Standard RBC gh
ost was included in each plate. After sample incubation, the plates were washed three times, with horseradish peroxidase (HRP) and monoclonal antibody YZ1.
(Changelian, PS et al., 1985, J. Immunol. 184: 1851-185.
8) Conjugates (Wilson, MB and Nakane, PK, 1978, Imm
unofluorescence and Related Staining Techniques, No
rth Holland Biomedical Press.pp.215-224)
S, diluted 1: 8000 in 50% blocking buffer, 100 μl
/ Well. After incubation at 37 ° C for 2 hours, the plates were washed again twice with PBS containing 0.05% Tween-20. Substrate o-phenylenediamine (OPH) substrate buffer (0.36% citric acid _{H 2 O, 1.74% Na 2} HPO 4 · 7H 2 O,
0.1% thimerosal, 0.4% H ₂ O ₂ , pH 6.3)
Added at 0μ / well. 2N H ₂ SO ₄ 50 after 20 minutes at room temperature
The reaction was interrupted using μ / well. The absorbance at 490 nm was read.

11.2. Gene modification of the CR1 coding sequence CR1 cDAN is composed of approximately 6,951 nucleotide base pairs (FIG. 1, sections 6 and 7 above). The translation termination signal for the native cDNA is located at base pair 6145. This protein is a membrane-bound receptor molecule composed of four long homologous repeats (LHR) exposed on the outer surface of the cell membrane, plus a membrane spanning domain of about 25 amino acids, followed by a carboxyl-terminal region extending into the cytoplasm. It is. This cytoplasmic domain is composed of 43 amino acids. Soluble CR1 molecule (sCR
The strategy used for the production of 1) was to remove the transmembrane region that anchors the protein to the cell membrane, and then express the truncated construct as a secreted polypeptide.

11.2.1. Format of pBSCR1c Plasmid pBSABCD (Example 8 above) contains the CR1 cDNA from nucleotides 1-6860 and has no EcoR V site at nucleotide 6860 from the untranslated sequence 3 '. CR
One cDNA has a unique restriction nuclease recognition site at base pair 5914, 29 bases from the start of the transmembrane domain. pBSABCD was first digested with Bal I to generate linear molecules with flash ends and then ligated using T4 DNA ligase to a synthetic oligonucleotide consisting of two 38 nucleotide complementary strands having the following sequences.

The resulting molecule had a restored Bal I site and an altered sequence regenerating the native CR1 sequence up to and including the alanine residue at the start of the transmembrane domain. In addition to this, a translation stop signal (in the following cases and underlined above) is inserted immediately after alanine,
Xh facilitates subsequent subcloning of altered cDNA
o There was an I restriction site.

 Xho I digestion of this plasmid (digested pBSCR1c)
Is the oligonucleotide-added Xho I site of the cDNA and CR1
 pBSKS + at the 5 'end of the cDNA Multiple cloning unit
CDNA insert by cutting at the Xho I site
(Digested sCR1c) was cut out. pBSCR1c has the following N-
Contains terminal sequence.

11.2.2. Formation of pBSCR1 A second sCR1 construct without a transmembrane region was generated as follows. pBSABCD was digested with SacI. This is CR
Unique Sac I at nucleotide base pair 5485 of one cDNA
Cleavage site and the Sac I site of the multiple cloning site of the host plasmid located at the 3 'end of the CR1 cDNA. This digestion was performed 1375 nucleotides from the 3 'end of the cDNA.
Excision of the DNA sequence occurred. This fragment was then removed by electrophoresis. The exposed ends of the resulting plasmid containing the remaining sCR1 cDNA were flushed with T4 DNA polymerase and blunt end ligation was performed. Pharmacia Translation Terminator (Catalog # 27-489)
Self-complementary oligomers containing translation termination signals in all three reading frames were also included in the ligation. Upon ligation, the inserted oligomer provided a new translation termination signal for the sCR1 cDNA.

11.2.3. Formation of pBM-CR1c pBMT3X is a eukaryotic expression vector containing the human metallothionein-1A gene (Krystal, M. et al., 1986, Proc. Natl. A
USA83: 2709-2713), which confers cells with resistance to increased levels of heavy metals such as cadmium. This vector also contains an engineered XhoI in front of the start codon for the Mt-1 protein.
Site containing the mouse metallothionein-1 gene. This Xho I site is used as an insertion site for gene expression under the control of the mouse Mt-I promoter.

The sCR1c insert (about 5.9 kb) was excised from pBSCR1c using XhoI and then ligated into the unique XhoI site of the vector pBMT3X. The correct orientation of cCR1c in pBMT3X was determined by restriction digestion (Maniatis, T. et al., 1982, Molecular Cloning,
A Laboratory Manual, Cold Spring Harbor Laboratory,
Cold Spring Harbor, New York). The resulting plasmid was named pBM-CR1c.

11.2.4. Deletion recombinant pT-CR1c1, pT-CR1c2, pT-CR1c
3. Formation of pT-CR1c4 and pT-CR1c5 Various deletion recombinants which specifically deleted the protein of sCR1 cDNA were also constructed (FIG. 20). Since each deleted recombinant lacked the transmembrane region of the full-length cDNA, expression of the recombinant would result in a soluble polypeptide.

11.2.4.1.pT-CR1c1 pBSCR1c digested with SmaI and sized 2.56 kb and 7.3 kb
Were generated. These fragments
Were separated by agarose gel electrophoresis.
Fragment was purified and ligated to itself. E.coli DH5α
Competent and non-competent cells (Morrison, D.A., 1979, Meth.
Enzymol. 68: 326-331) and then transformed with the ligation mixture
did. The resulting plasmid was named pBL-CR1c1. This
Constructs are 38% LHR-B and 100% LHR-C of the CR1c insert.
And 51% of LHR-D were removed. In addition, this is
Regenerate the Sma I site at 5/4894 bp, retaining the correct translational frame
I was pBL-CR1c1 is digested with XhoI and the CR1 insert is digested with pBle
uscript Isolated from vector. Isolated CR1 flag
The unique Xho I site of the expression vector pTCSgpt
Into plasmid pT-CR1C1.

11.2.4.2.pT-CR1c2 Digestion of pBSCR1c with Cla I and Bal I 3.96 kb in size
And two fragments of 5.9 kb were generated. these
Was purified from an agarose gel. plus
Mid pBR322 was digested with Cla I and Bal I and 2.9 kb pBR322
The fragment was purified and the 5.9 kb fragment from pBSCR1c
Connected to Transform E. coli DH5α cells with the ligation mixture
The resulting plasmid was designated as pBR8.8. This
Rasmids are digested with Xba I and are 7.45 kb and 1.35 kb in size.
Were generated. 7.45kb fragment
Was purified from an agarose gel and ligated to itself.
Digest the resulting plasmid pBR7.45 with Cal I and Bal I
And an isolated 4.5 kb fragment containing the sCR1 cDNA
Was ligated to the 3.96 kb fragment from pBSCR1c, plus
Mid pBL-CR1c2 was generated. This construct is 90% LHR-B
Was removed in the sCR1 insert and the junction 1637 / 2987p
The Xba I site was regenerated with b and retained the correct reading frame. pB
Digest L-CR1c2 with XhoI and insert the sCR1 insert into pBleuscript.
Isolated from vector. The isolated sCR1 fragment
It is then inserted into the unique Xho I site of the expression vector pTCSgpt
As a result, a plasmid pT-CR1c2 was generated.

11.2.4.3.pT-CR1c3 pBSCR1c was digested with NsiI and
And 7.46 kb of three fragments. 7.46kb hula
Purified from agarose gel and ligated to itself
This produced the plasmid pBL-CR1c3. This construct
77% of LHR-A and the remainder of the CR1 insert had been removed.
The Nsi I site was regenerated at junction 463/2907 pb. Nonsense
Introduce a codon immediately after the regenerated Nsi I site
Was used to modify the translation frame. Digest pBL-CR1c3 with XhoI
And insert the sCR1 insert into pBleuscript Isolated from vector
Was. The inserted sCR1 fragment is then transformed into an expression vector
Inserted into the unique XhoI site of pTSCgpt, the plasmid pT-CR
1c3 was generated.

11.2.4.4.pT-CR1c4 pBSCR1c was digested with PstI. pBleuscript Polyli
The Pst I site of the linker contains the CR1 cDNA
(See Example 8.1 above). Generate
1.35 kb and 8.5 kb fragments
Separation by electrophoresis and purification of the 8.5 kb fragment.
Ligation to itself generated the plasmid pBL-CR1c4. This
Constructs are 31% of LHR-A and LHR-B of the sCR1 insert.
69% were removed. Regenerate Pst I site at junction 1074 / 2424pb
And thus retained the correct reading frame. pBL-CR1c4
Digest with Xho I and insert the sCR1 insert into pBleuscript Vector
Separated. The isolated sCR1 fragment is then expressed
Insert the unique Xho I site of the vector pTCSgpt into plasmid
PT-CR1c4 was generated.

11.2.4.5.pT-CR1c5 pBL-CR1c1 was digested with SmaI, thus
Was linearized at the unique Sma I site. This plasmid
Dephosphorylated nonsense codon (New England Bi
Phosphorylated Nhe I phosphorus containing olabs, Beverley, MA)
Connected to the car. This type of linker has all three
Contains a translation stop codon in a possible reading frame and has an Nhe I restriction
Also contains a nonsense linker in the sCR1 cDNA.
Make it easy to confirm the existence of a key. Exclude pBL-CR1c5 with XhoI
And sCR1 insert into pBleuscript Isolated from vector
Was. The isolated sCR1 fragment is then transformed into an expression vector
Plasmid pT-CR inserted into the unique XhoI site of pTCSgpt
1c5 was generated.

11.3. Expression of Soluble CR1 As shown here, the expression of a soluble form of CR1 that can be secreted from cells in high yield depends on (i) one of the CR1 DNAs used in the CR1 DNA used for deletion or truncation. It is not limited to position, and (ii) is not limited to the use of special expression vectors (see below). The ability to produce secreted sCR1 has been demonstrated in two different expression systems.

11.3.1. Formation of the pTCS Series of Expression Vectors The pTCS series of expression vectors used consisted of three plasmids, each with a unique plasmid for cDNA insertion.
It has an Xho I cloning site (FIG. 21). Transcription of the inserted cDNA is driven by a set of tandem promoters. The SV40 early promoter is located upstream of the adenovirus 2 major delayed promoter (AD2 MLP). cDNA
There is an adenovirus tripartite leader between the beginning of AD2 and the AD2 MLP. The transcribed mRNA is terminated by a polyadenylation signal provided by a mouse immunoglobulin kappa (Igκ) sequence located downstream of the Xho I cDNA cloning site. Xanthin-a selectable marker
Guanine phosphoribosyltransferase (gpt),
Dihydrofolate reductase (dhfr) or neomycin resistance (neo ^r) respectively pSV2gpt, pSV2dhfr, or pSV2n
Supplied with the insertion of the corresponding marker from eo. These plasmids were also the source of the bacterial system for replication and beta-lactamase genes for ampicillin resistance. In general, the choice of which of these vectors to use depends on which selectable marker or marker combination is preferred for recombinant selection.

 The complete DNA sequence is adenovirus 2 (AD2), SV40, pS
V2cat (Gorman, C., 1985, DNA Cloning, Volume II, A Prac
tical Approach, ed.D.M.Glover, IRL Press, pp. 143-19
0) and known about mouse immunoglobulin kappa
ing. Sequence is GenBank Database and National
In Biomedical Research notes and references.
Either of these sequences must be in the appropriate segment of the pTCS vector.
Serving as a source for

The vectors pTCSgpt, pTCSneo and pTCSdhfr were formed from the intermediate plasmids pEAXgpt and pMLEgpt as follows.

11.3.1.1. Formation of pEAXgpt Step 1. Ad2 MLD NDA fragment was derived from M13 mp9 / MLP (Concino, MF et al., 1983, J. Biol. Chem. 258: 8493-8).
496). This plasmid contains a P at nucleotide 6069.
Sac II at vu II restriction site and nucleotide 5791
Nucleotides 5778 (Xho I site) to 6231 including the site
(Hind III site) containing the adenovirus 2 sequence (see NBRF Nucleic database, accession # Gdad2). X
ho I or Hind III fragment was converted to Hind III of M13 mp9.
And the plasmid M13 m
p9 / MLP was generated.

Plasmid M13 mp9 / MLP was digested with EcoR I and Hind III and smaller MLPs containing the fragment were isolated. pUC plasmid (Pharmacia, Inc., Piscatway, NJ)
Digest with EcoR I and Hind III and remove the larger fragment from this plasmid then EcoR I to Hind III
Ligation to the MLP fragment. This generated a novel MLP-containing plasmid with the plasmid backbone of pUC. This plasmid was digested with Sma I, ligated to a Sal I linker and circularized again. The new plasmid was then digested with Pvu II, which cleaved the plasmid at the Pvu II site located at position # 6069 within the adenovirus 2 insert. The resulting linear fragment was ligated to an Xho I linker and circularized again. This plasmid is then
Upon digestion with I and Sal I, a smaller fragment containing MLP DNA was isolated (fragment # 1).

Step 2. Plasmid pSV2gpt (American Type Culture Col
Section (ATCC) No. 37145) is digested with Pvu II and Sal
It was ligated to I linker and digested with Sal I. The final product was a linear pSV2gpt fragment that served as the gpt gene source (fragment # 2).

Step 3. Mouse immunoglobulin Igκ fragment (Hiete
r, PA et al., 1980, Cell 22: 197-202) for Hind III and A.
Digestion with va II and isolation of the fragment containing the polyadenylation sequence. NBRF Nucleic Database (Contract #
mouse Ig kappa sequence available at κ cms) has an Ig stop codon at position 1296, followed by an Ava II site at 1306, 14
There is an AATAAA polyadenylation site at 84 and a Hind III site at 1714. Insert the overhanging end of this fragment into E.co.
The fragment was ligated with XhoI linker and digested with XhoI. This fragment (fragment # 3) served as the source of the polyadenylation site.

Step 4. Fragments 1, 2 and 3 were ligated together by T4 DNA ligase to generate a circular plasmid. The correct orientation of the fragments in this plasmid was confirmed by restriction analysis. Downstream of the Xho I cDNA cloning site was a mouse kappa polyadenylation site, and further downstream from this site was the SV40 promoter and gpt gene. Upstream of the Xho I site is the MLP promoter,
Further upstream from this promoter were bacterial replication and ampicillin genes. This plasmid was then digested with Sal I and the overhang ends were filled in with E. coli DNA polymerase. The resulting blunt end fragment was ligated to EcoRI linker and recircularized with T4 DNA ligase. This final plasmid was named pEAXgpt.

11.3.1.2. Formation of pMLEgpt Step 1. Plasmid pMLP CAT (Lee, RF et al., 1988, Virolog
y, 165: 50-56) is an expression plasmid with a pML vector background, adenovirus 2MLP and CAT
Contains the 3 'segmented leader sequence 5' to the gene. pM
LP CAT was digested with Xho I and Sac II; Xho I cut at the site between the CAT gene and the L3 region of the tripartite leader, and Sac II was located at position # 5791 in adenovirus DNA, except for MLP Cut at 5 '. Thus, the AD2 MLP and tripartite leader were located on this small Xho I up to the Sac II fragment (fragment # 4).

Step 2. Plasmid pEAXgpt was digested with Xho I and Sac II and the smaller MLP containing fragment was discarded.
A larger fragment (fragment # 5) was isolated. Fragments 4 and 5 having both Sac II and Xho I ends were ligated to generate plasmid pMLEgpt.

11.3.1.3. Formation of pTCSgpt Step 1. pMLPEgpt was digested with SacII and the ends were filled in with T4 DNA polymerase to generate a blunt end fragment (fragment # 6). This Sac II site is located 5 'of the MLP-3 segmental leader at nucleotide 5791 in the adenovirus 2 sequence.

Step 2. Transfer pSV2dhfr (ATCC Accession No. 37146) to Hind III and
Digested with Pvu II. The smaller 342 nucleotide fragment containing the SV40 early promoter was
A blunt end was obtained using the Klenow fragment of the polymerase (fragment # 7). Fragments 6 and 7 were ligated with T4 DNA ligase. Restriction enzyme analysis showed that these fragments were oriented correctly and the Xho I cD
Two tandem promoters were provided upstream of the NA cloning site, and it was confirmed that each promoter could induce RNA synthesis in the same direction. This plasmid is called pTCSgpt
(Fig. 22).

11.3.1.4. Formation of pTCSdhfr Step 1. pSV 2 dhfr was digested with Hind III and Pvu II, and the larger fragment was purified from an agarose gel (fragment # 8). The smaller SV40 early promoter containing fragment was discarded.

Step 2. pTCSgpt was digested with EcoRI and then filled in with the Klenow fragment of E. coli DNA polymerase to generate blunt ends. This linear fragment is then
Digested with II, pTCS transcription unit of SV40 promoter, ML
P, three-segment leader, Xho I cDNA cloning site,
A fragment (about 1600 nucleotides) containing the mouse Igλ sequence and a second SV40 promoter was isolated (fragment # 9). This fragment had one flash end and one Hing III overhang end. The ligation of fragments 8 and 9 was performed using plasmid pT
CSdhfr was generated.

11.3.1.5 Formation of pTCSneo Step 1. Convert pSV2neo (ATCC No. 37149) to Hind III and BamH
Digest with I and use the larger fragment (fragment #
10) was isolated. This fragment contained the plasmid background and the neo gene.

Step 2.Digest pTCSdhfr with Hind III and BamHI, pTCSdhfr
The transcription unit (fragment # 11) was isolated from an agarose gel after electrophoresis of the digestion product. Fragment
Ligation of 10 and 11 produced plasmid pTCSneo.

11.3.2. Expression and Assay of Plasmids pBSCR1c, pBSCR1s and pBM-CR1c, Mammalian Expression Vector Containing Soluble CR1 Cloning Sequence 11.3.2.1. CR1 Truncated at Different Positions in CR1 cDNA
Expression of constructs Plasmids pBSCR1c and pBCSR1s are formed (see
Section 11.1), except for the transmembrane and intracytoplasmic regions
The cDNA coding region was preserved (Fig. 20). pBCS
R1s is the SCRs2 present in a portion of LHR-D and pBSCR1c
Shorter than pBSCR1c because it lacks 9 and 30. The sCR1 portion of these plasmids was inserted into pTCSgpt and then transfected and expressed as described above.

pBSCR1c / pTCSgpt construct: pBSCR1c digested with Xho I to 5.9k
The b insert generated sCR1c. sCR1c to pTCSgpt Xho I cD
Insertion into the NA cloning site generated pBSCR1c / pTCSgpt.

pBSCR1s / pTCSgpt construct: pBSCR1s was digested with XhoI and PvuI to release the sCR1s insert. Both ends of the insert were blunted with T4 DNA polymerase. This insert was purified from an agarose gel. Vector pTCSgpt for Xho I
And the overhang Xho I end was filled in with E. coli DNA polymerase I. The sCR1s insert was then ligated into the blunt end vector to generate pBSCR1s / pTCSgpt.

Plasmids pBSCR1c / pTCSgpt and pBSCR1s / pTCSgpt
The linear DNA produced by digestion with Fsp I was transfected into Chinese hamster ovary cells, which were transformed with dh by calcium phosphate coprecipitation with plasmid pSV2dhfr.
It was a recombinant in the fr gene (CHO DUX B11 cells).
Transfectants were selected for their ability to grow in DHFR selective media. Culture supernatants of transfectant clones were assayed for selected sCR1 by ELISA. Culture supernatants from 50 pBSCR1c / pTCSgpt recombinants were assayed and positive recombinants were removed by a growth method by culture in increasing concentrations of methotrexate. In addition, pools of transfectants were created by culturing together eight pBSCR1c / pTCSgpt transfectants for each pool and carrying them by the same amplification method. The results of the amplification are shown in Table VI.

Twelve recombinants from pBSCR1s / pTCSgpt were assayed for production of soluble CR1 by ELISA. All 12 candidates showed appreciable levels of secreted sCR1. The best producers produced sCR1 levels comparable to those produced by the best pBSCRc1 / pTCSgpt transfectants.

The pBSCR1c / pTCSgpt and pBSCR1s / pTCSgpt recombinants produced soluble CR1 at similar production levels. This is soluble C
The ability to produce R1 polypeptide was shown to be independent of its truncated point in the CR1 cDNA.

The first attempt to amplify a cell line, clone 35.6 beyond 500 nm methotrexate was unsuccessful. For this reason, other cell lines were sought from the panels shown in Table VI. Candidates were selected based on expression corresponding to clone 35. The system was subcloned and screened for sCR1 concentration on the supernatant, with methotrexate levels at which expression began to plateau. Cell lines 15 resistant to 50 nm methotrexate
Was chosen based on this. It was subcloned to create the cell line 15.19 and increased expression to 35.6 levels.
15.19 was then cultured in media containing various concentrations of methotrexate. Only 2500nm methotrexate has sCR
This resulted in increased expression levels, increasing secretion by 63% from ± 35.6. This cell line, named 15.192500, was then subcloned by limited dilution and the cell line 15.192500.0
Created 7,15.192500.10 and 15.192500.65. These cell lines have sCR1 per ml of 143, 119,
And 103% more.

11.3.2.2. Expression of sCR1c in two different expression systems The truncated CR1 cDNA insert, sCR1c, is used as an expression vector pTCSg
Inserted into pt and expressed as described above. Also, make it one
Insertion into the expression vector pBMT3X as described in section 1.2.3 resulted in pBM-CR1c. Both of these expression vectors have very strong promoters. To determine whether one system produced a better product of the secreted polypeptide, the expression of soluble CR1 was tested in both systems.

C127 I mouse cells (ATCC accession number CRL1616, Rockvill
e, Maryland) were transfected with pBM-CR1c using the calcium phosphate method (Graham, FL and van der Eb, AJ, 1973, Virology 52, 456-467). After glycerol shock, cells were harvested with 10% fetal bovine serum and 2 mM
D-MEM medium containing L-glutamine was supplied again.
Incubated at 48 ° C for 48 hours. The cells were then trypsinized and split at 1: 5 and 1:10 in complete D-MEM medium + 10 μM cadmium chloride. Cadmium resistant colonies appeared within 10 days. Ten colonies were picked by using a cloning cylinder. Each colony was transferred to a 60 mm Petri dish containing complete D-MEM medium,
The cells were incubated at 37 ° C., 5% CO ₂ until the cells reached confluence. Thereafter, for each dish, the cells were trypsinized, divided into three 60 mm dishes, and frozen cell line preparation, RN
A extraction, used for ELISA test of cell culture medium for the presence of secreted sCR1c.

When the cell media was removed from each confluent petri dish and subjected to ELISA analysis, all pBM-CR1c clones tested were positive for soluble CR1 production. The level of secreted sCR1 from the pBM-CR1c recombinant was comparable to the amount secreted from the pBSCR1c / pTCSgpt recombinant. This indicated that the ability to produce high levels of secreted sCR1 polypeptide was not dependent on the use of certain promoters or expression systems alone.

11.3.3. Mammalian expression vector containing soluble CR1 coding sequence, plasmids pT-CR1c1, pT-CR1c2, pT-CR1c
3, Expression and analysis assays of pT-CR1c4 and pT-CR1c5 The pT-CR1c series of deletion mutants, like the components pBSCR1c and pBSCR1s, lost the transmembrane and cytoplasmic domains. In addition, the deletion mutants also contained rather large deletions of various LHR regions of the CR1 cDNA (see FIG. 20).
Deletion mutants were expressed in CHO DUX B11 cells and the level of soluble CR1 polypeptide produced was measured.

For each deletion construct, 40 different pools of clones were selected for ELISA analysis to determine whether soluble CR1 polypeptide was being produced. 5 pT-CR1
c All constructs were determined by either ELISA or the presence of functional activity in the cell media and were found to secrete sCR1 into the cell media. Supernatants from cells transfected with four of the five pT-CR1c constructs have functional sCs as determined by a hemolysis assay.
R1 was produced (see Table VII and Section 13.2 below).

The fact that the deletion mutants were also able to produce soluble CR1 further showed that the ability to express sCR1 did not depend on one precise genetic modification of the CR1 cDNA. All constructs were able to produce soluble polypeptide as long as the transmembrane region was deleted.

12. Example: Production and purification of soluble CR1 Large quantities of sCR1 were produced in a hollow fiber bioreactor system. The amount of sCR1 obtained was proportional to the relative yield of the inoculated recombinant clone. For optimal purification results, serum-free media was selected that resulted in high production of sCR1 in the absence of large amounts of externally added fetal calf serum polypeptide.

12.1. Large-scale production of soluble CR1 Cell-Pharm, a trademark, cell culture system I (CD Medical Inc. Miami Lakes, FL) equipped with a model IV-L hollow fiber bioreactor (30 kD molecular weight cutoff) under aseptic conditions Assembled. Two clones (clone 2 and clone 35 of pBSCR1c / pTCSgpt) were grown in eight T-225 flasks. At confluence, trypsinize the bacteria, wash,
Pelleted and resuspended in medium. About 5 of clone 2
× 10 ⁸ cells and about 10 × 10 ⁸ cells of clone 35 were inoculated into two separate hollow fiber bioreactors. Alpha MEM + 10% fetal calf serum, 8 mM L-glutamine, 100 μg / ml penicillin-streptomycin and appropriate concentration of methotrexate (50 for clone 2)
nM; 500 nM for clone 35) of 20 culture reservoirs were used. The mixed gas (5% CO _{2 in} air) was blown into the reservoir medium with an oxygenator to maintain the pH. Medium circulation rate, exchange rate and gas flow rate were adjusted to produce maximum production. Samples were collected by inoculum, centrifuged at 1000 rpm for 10 minutes, filtered through a 0.22 μM pore size filter, and kept at 4 ° C. before purification. The recovery volume and frequency were gradually increased from 25 ml, 3 times a week at the start of culture, to 40 ml, 5 times a week after 2-3 crops. CR sCR1 production
Assayed by 1 ELISA. In the first month after inoculation, the yield of clone 2 and clone 35 was 66 μg / day and 66 μg / day, respectively.
It was 1060 μg / day. These yields increased as the culture became established.

12.1.1. Production of sCR1 in serum-free media Two commercially available serum-free media were used for cell growth and sC1
Tested for their ability to support the production of R1. pB
Two confluent T75 flasks of SCR1c / pTCSgpt clone 35
Divided into 5 flasks. One flask was cultured in alpha MEM supplemented with 10% fetal calf serum, L-glutamine, antibiotics, and 500 nM methotrexate. Another flask was filled with alpha MEM + L-glutamine, antibiotics, 500 nM methotrexate + HB CHO growth supplement (Han
a Biologics Inc. Alameda, CA)
From 1%, 0.5% and 0%. The cell growth and sCR1 production of the two flasks were compared. Growth of cells in medium without serum did not reach confluence.
The levels of sCR1 production are shown in Table VIII. In each case, sCR1
The level of production was best when cells were grown in 10% fetal calf serum. For comparison, the level seen at day 14 in serum-free medium was 1. compared to 4.2 x 10 ¹⁰ ghosts / ml for medium supplemented with 10% fetal calf serum.
4 × 10 ¹⁰ ghosts / ml.

Cell growth and sCR1 production in recombinants was determined using a second source of serum-free medium (CHO-1, Ventrex Laboratories Inc.
(Portland ME). Since it was not necessary to wean the cells grown in serum into this medium, the cells were thawed and cultured directly in medium without serum. This medium consists of a DME-F12 base and a growth additive. An equal number of cells were thawed and seeded in separate wells of a 24-well plate. After the cells attach, discard the medium,
Medium containing 10% fetal calf serum or medium without serum was added to appropriate wells. Each condition was duplicated. Unlike the serum-free medium tested earlier, CH
O-1, the Ventrex Laboratories medium, produced similar amounts of proliferating cells as the medium containing fetal calf serum.

12.1.2. Conclusions The results above indicated that CHO cells producing sCR1 could be maintained in media without certain sera. this is,
This has resulted in a reduction in the cost of the medium for large-scale production experiments. Yet another advantage was that purification of sCR1 from cell culture supernatant was simplified. This was because there was no need to remove fetal calf serum proteins.

12.2.Purification of soluble CR1 With the advent of specific anti-CR1 antibodies, the number of chromatographic steps required to produce purified CR1
It has become possible to replace with a simplified two-step operation. This increases the yield of CR1 protein that can be obtained.
Won increased to about 1-5 mg CR1 per 5.9 x 10 ¹³ red blood cells
g, WW et al., 1985, J. Immunol. Methods 82: 303-313). However, it was always necessary to solubilize the CR1 containing material in detergent since the reported purification was CR1 in membrane bound form.

Soluble produced by recombinant transfectants
CR1 does not need to be solubilized with a detergent for purification. It is already soluble. Although soluble CR1 can be purified by anti-CR1 antibody chromatography (see below), this procedure does not lend itself to large-scale production. The extent of scale-up is limited by the amount of anti-CR1 antibody that can be obtained to prepare an antibody matrix for an antibody purification column. In addition, the high binding affinity of antibodies such as YZ-1 for CR1 can be attributed to rather harsh conditions, such as pH 12.2.
Wong, W., means that sCR1 product needs to be used to remove the bound sCR1 product from the antibody matrix.
W., 1985, J. Immunol. Methods 82: 303-313).

Due to the ability to purify very large amounts of soluble CR1, a purification procedure involving an HPLC column was developed. These HPLC columns can be easily scaled up to produce larger quantities of purified soluble CR1. In addition, they do not require harsh conditions for elution and recovery of sCR1.

12.2.1. Antibody affinity column purification 12.2.1.1.MethodFor antibody affinity purification of sCR1, 100 mg of monoclonal antibody YZ-1 was prepared according to the manufacturer's instructions according to the manufacturer's instructions.
-10 BioRad, Richmond, CA). The supernatant containing CR1 from cell culture was incubated overnight at 4 ° C. with immobilized YZ-1 in a rocking flask. The material was injected into a glass column and 10 mM Hepes, 0.1 M NaC
l, thorough washing at pH 7. Yoon, SH eluted sCR1 with 20 mM sodium phosphate, 0.7 M NaCl, pH 12.
And Fearon, DT, 1985, J. Immunol. 134, 3332-3338).
The eluted fractions are collected using the Biorad Protein Assay (Biorad®
ichmond, CA). Collect the protein-containing sample directly and use 0.1M Hepes pH
Dialyzed overnight at 4 ° C. in (2 × 1). Thereafter, the sample was dialyzed in PBS. The presence of sCR1 was analyzed by CR1 ELISA.

12.2.1.2. Results The cell culture supernatant containing sCR1 produced by the transfectant pBSCR1c / pTCSgpt clone 2 was loaded on an anti-CR1 antibody affinity column, and the peak sCR1 fraction was collected. A portion of this purified material was applied to a 4-20% SDS-PAGE gel on DAIICHI Inc .; polyacrylamide gel; Laemmli
UK modification, 1970, Nature 227, 680-685). Under reducing conditions, the apparent molecular weight of soluble CR1 was approximately 224,000 daltons (FIG. 24). This purified CR1 was also shown to be active due to its ability to inhibit complement-mediated hemolysis and C5a and C3A production (13, below).

12.2.2.CR1 purification by HPLC 12.2.2.1.Methods 12.2.2.1.1.Starting material When a culture was first established in a bioreactor, the level of sCR1 production was determined if the culture had grown for several months. Less than if you had. Generally, by the time the cells in the bioreactor reach confluence and produce the maximum amount of sCR1,
There was a period of several weeks. Cell culture supernatants containing small amounts of sCR1 could be concentrated by ammonium sulfate precipitation or ultrafiltration before purification. Ammonium sulfate fractionation of the supernatant, ranging from 60-80% saturation, precipitated sCR1 in substantially equal yields. Dissolve the precipitate to the minimum volume and use cation exchange HP
Dialyzed in starting buffer for LC. The CHO cell culture supernatant could also be concentrated by ultrafiltration and dialyzed into starting buffer for cation exchange chromatography.

As the bioreactor produced higher concentrations of soluble CR1, CHO cell culture supernatants from these cultures could be dialyzed directly into starting buffer for cation exchange chromatography.

12.2.2.1.2. Cation exchange HPLC procedure Samples were run in starting buffer (0.02M sodium phosphate, 0.06N
Dialysis in sodium chloride, pH 7.0), followed by 0.2 μl
m and filtered to remove particulate matter. afterwards,
The sample was loaded on a cation exchange high performance fluid chromatography column (10 cm × 10 mm, Hydropore-SCX HPLC column from Rainin). The column was washed, eluted with a sodium chloride gradient and eluted with 0.02M phosphate, 0.5N NaCl, pH 7.0. sCR1 eluted somewhere between 0.06N and 0.25N NaCl. Elution was monitored by absorbance at 280 nm and ELISA.

12.2.2.1.3. Anion Exchange HPLC Procedure If desired, further purification of the cation HPLC purified sCR1 could be obtained by anion HPLC. The peak fraction from anionic HPLC was dialyzed into starting buffer for anionic HPLC. The sample was loaded and the column (Hydropon from Rainin)
oro-AX) was washed in a 0.01 M phosphate solution, pH 7.5. The column was eluted with a series of steps and gradients using 0.01 M phosphate, 0.5 N NaCl, pH 7.5. sCR1 is 0.0
Eluted somewhere between N-0.3N NaCl. Elution was monitored as described above for cation exchange HPLC. The concentration and pH of the cation HPLC column buffer and the anion HPLC column buffer are only exemplary. Other buffer concentrations, salt conditions, or pH conditions are also possible.

12.2.2.1.4. Western blot analysis Towbin, H. et al., 1979, Proc. Natl. Acad. Sci. USA, 76, 4350.
Western blotting was performed using the modified procedure from -4354. Briefly, 4-20% of purified sCR1
Treated by SDS-PAGE, transferred to nitrocellulose, anti-CR1
(Mouse mAb YZ-1 or J3D3) and detected with goat anti-mouse antibody conjugated to alkaline phosphatase.

12.2.2.2. Results For a typical experiment, 50-100 ml of the supernatant from the bioreactor culture was dialyzed into starting buffer and loaded on a 10 cm x 10 mm cation exchange HPLC. ELISA and peak fraction
Measured by absorbance at 80 nm and pooled. The protein concentration of the pool was measured by absorbance at 280 nm (estimated from the CR1c amino acid composition, ε at 280 nm
(1%) = 10). Dozens of mg were purified from 100 ml of the amplified culture supernatant.

As an example, the transfectant pBSCR1c / pTCSgpt
100 ml of culture supernatant from clone 2 produced 22 mg of purified sCR1 when purified by cationic HPLC as measured by absorbance at 280 nm (FIG. 24). CR1
When monitored by ELISA, the yield was calculated to be 202%, with the other 13% being in the through or column wash fractions. Yields greater than 100% probably reflect matrix action in the ELISA.

Given the rate at which culture supernatants can be withdrawn from the bioreactor, producing about 100 mg of purified soluble CR1 per bioreactor per week should be possible with this level of methotrexate amplification It is. Some ways that this level of production can be scaled up include maximizing the starting culture with methotrexate before inoculating the bioreactor, increasing the number of bioreactors in production at any one time. , And using a large volume HPLC column.

12.2.2.3. Characterization of purified soluble CR1 Peak fraction containing sCR1 from cationic HPLC (Figure 24)
Was further purified by anionic HPLC. In various processes
The purity of the sCR1 material was tested by SDS-PAGE (25th
Figure). The smaller band seen in these heavily loaded gels is the anti-CR1 monoclonal antibody YZ1 or J3
Representative of a fragment of sCR1 as determined by Western blot analysis using D3. Fragment sC
The R1 band was not found in most preparations.

The functional activity of purified sCR1 was tested by its ability to inhibit 50% of the main pathway complement mediated hemolysis at a purified sCR1 concentration of 0.25 μg / ml. In addition, purified soluble CR1 can inhibit 50% of the main pathway complement C5a production at 5 μg / ml and 50% of C3a production at 13 μg / ml.
Could be suppressed (see paragraph 13 below).

12.2.2.4. Conclusion As noted above, we have determined that sCR1
An improved method for the purification of soluble CR1 has been developed that can be easily scaled up to produce high amounts of. The basic element of this operation is already soluble and thus bound to the membrane
It contained a starting material that obviated the need to solubilize CR1 with detergent. The reduction of fetal bovine serum concentrations in bioreactor cultures and / or the use of alternative media in these cultures eliminates the need to remove high levels of foreign protein from starting materials, including sCR1, during subsequent purifications. did. In addition, the development of HPLC procedures for purification has provided a large-scale purification method. Cation HPLC or a combination of cation HPLC followed by anion exchange HPLC may be used for purification. High yields of substantially pure soluble CR1 can be obtained in this operation in only one or two steps.

13. Examples: Demonstration of in vitro activity of soluble CR1 13.1. Inhibition of neutrophil oxidative burst Reperfusion of tissue damage incurred during myocardial infarction
n) In the injury model, active complement components induce neutrophil adhesion and activation. Activated neutrophils undergo an oxidative burst that produces highly toxic oxygen radicals. These and other potential toxins are released during neutrophil granulation and damage surrounding tissues. Soluble CR1 contains C3a and C5
a, the area of damaged tissue may be reduced by preventing the generation of complement components involved in neutrophil activation.

To monitor the ability of soluble CR1 to block the development of C5a during complement activation in vitro, Bass, using a bioassay that can quantify the generation of oxygen radicals generated by neutrophils in a C5a-induced oxygen burst. DA et al., 1983, JI
mmunol. 130, 1910-1917). This analysis uses dichlorofluorescein diacelate (DCFDA), a lipid-soluble molecule that can be trapped inside cells and become highly fluorescent upon oxidation.

13.1.1. Materials and Methods 13.1.1.1. Materials New whole blood, human complement source (Beth Israel Hospital, Bos)
ton, MA), dried baker's yeast, PBS containing 0.1% gelatin and 5 mM glucose, 100 mM EDA, HBSS (Kodak)
10 mM DCFDA, a buffer (Ortho) to lyse red blood cells (RBC)
Diagnostics, purified C5a (Sigma Chemical Co., St. Louis,
MO), and soluble CR1.

13.1.1.2. Preparation of Neutrophils Neutrophils were prepared as described by Bass (198
3, J. Immunol. 130, 1910-1917). 2.0 ml of whole blood was washed three times in PBS-gelatin-glucose and 10 μM DC in HBSS
Resuspend in 5 ml of FDA + 5 ml of PBS-gelatin-glucose,
Incubated at 37 ° C for 15 minutes. Thereafter, the cells were centrifuged and PBS-gelatin-glucose + 5 mM EDTA 2.0 ml
In water.

13.1.1.3. Preparation of yeast particles Dried baker's yeast was resuspended in water, washed twice and boiled for 30 minutes. Re-wash the particles twice in water, 0.5g / m in water
l (Simpson, PJ et al., supra).

Activation of neutrophils with purified C5a 100 μl of cells containing DCFDA were treated with a buffer that dissolves RBC, washed once in PBS-gelatin-glucose-EDTA, PBS-gelatin-glucose Resuspended in 1.0 ml. 200 ng / ml purified C5a or control 50μ with target cells 0.5
ml at 37 ° C. and analyzed by float cytometry at various time intervals.

13.1.1.5 Purified C5a in human serum or plasma
Activation of neutrophils by DCFDA-loaded cells 100 μl diluted 1: 1 with human serum or heparinized plasma (100 ng / ml) or control
Incubated in 50μ of 5a at 37 ° C for 30 minutes. RB
C's were lysed and neutrophils were analyzed on a flow cytometer.

13.1.1.6. Activation of neutrophils by human serum or plasma activated by yeast particles New frozen serum and plasma 425μ + sCR1 or control 50
μ was incubated for 30 minutes at 37 ° C. with 25 μ of yeast particles. Thereafter, the complement activation sample and the control sample were centrifuged to remove yeast particles. Each of these samples
Ten two-fold dilutions were made in PBS-gelatin-glucose-EDTA. 50 μl of serial dilutions of control and activated serum and plasma were added to 50 μl of target cells containing DCFDA.
And incubated at 37 ° C. for 30 minutes. Thereafter, PBC was dissolved and removed, and neutrophils were analyzed by flow cytometry.

13.1.2.Results 13.1.2.1.C5a induces an oxygen burst in human neutrophils that can be measured using DCFDA. FIG. 26 shows the rapid increase in fluorescence intensity of human neutrophils after stimulation with purified C5a. Shows a significant increase. Within 4 minutes of addition of C5a (20 ng / ml final concentration), neutrophils were 10 times brighter than control DCFDA-containing neutrophils. By 20 minutes, neutrophils should be
It was twice brighter. This analysis appears to be a sensitive indicator of C5a.

13.1.2.2.Human serum contains DCFDA, which blocks the oxygen burst effect of purified C5a on neutrophils, and neutrophils incubated with purified C5a diluted in human serum increase fluorescence intensity Was not observed. This effect may be due to platelet derived growth factor (PDGF) released from platelets during coagulation. It has been shown that small amounts of PDGF can suppress C5a-induced neutrophil activation (Wilson, E. et al., 1987, Proc. Natl.
Acad. Sci. USA 84: 2213-2217).

13.1.2.3. Heparinized plasma does not block the effect of C5a on neutrophils C5a diluted 1: 1 in heparinized plasma induced an oxygen burst in neutrophils containing DCFDA. In buffer
Although not as significant as C5a, there was a 10-fold increase in fluorescence intensity after 30 minutes incubation with neutrophils. Decreased signal can be caused by PDGF release during phlebotomy or plasma separation. A milder and faster unit of plasma from the cellular components of the blood can minimize the release of PDGF and result in better C5a function.

SCR1 present during complement activation blocks C5a developmentZymosan-induced activation of human complement in the presence of soluble CR1 showed reduced C5a activity as measured by DCFDA analysis . As can be seen in FIG. 27, a 1:16 dilution of human plasma activated in the presence of sCR1 was more likely in neutrophils than a 1:16 dilution of plasma activated in the absence of sCR1. A 70% decrease in fluorescence intensity resulted. This means suppression of C5a generation by sCR1. Further optimization of the DCFDA assay and plasma collection will result in a more dynamic and sensitive analysis of soluble CR1 activity.

13.2. Inhibition of complement-mediated hemolysis 13.2.1. Methods Complement-mediated erythrocyte lysis (hemolysis)
Was tested by analyzing for inhibition of Inhibition of hemolysis was measured as a function of soluble CR1 concentration. Transfer the sCR1 sample to be tested to 0.1 M Hepes buffer (0.15 N NaCl, pH 7.4)
And 50 μl was added to each well of the V-shaped bottom microtiter plate, typically in triplicate. The human serum used as a complement source was diluted 1: 125 in Hepes buffer and 50μ was added to each well. Next, commercially available sheep erythrocytes (Di-
amedix catalog number 789-002) was used as received and added at 100 μ / well to initiate the complement pathway to hemolysis. Plates were incubated at 37 ° C. for 60 minutes, followed by centrifugation at 500 × g for 10 minutes. The supernatant was removed and placed in a flat-bottomed microtiter plate. The extent of hemolysis was measured as a function of the absorbance of the sample at 410 nm. Maximum absorbance (corresponding to highest hemolysis) Amax
Is the absorbance As- of a red blood cell sample containing only human serum.
It was obtained from the absorbance Ao of a sample containing only erythrocytes. Therefore, Amax = As−Ao. The difference between the absorbance of a red blood cell sample containing both human serum and sCR1 and the absorbance of a cell sample containing only sCR1 was defined as A sample. Inhibition represents an IH as a fraction (Amax-A sample / Amax), was defined as the concentration of sCR1 required to produce a value of the IH ₅₀ IH = 1/2. To monitor the chromatographic fractions, no serum-free control was included and anti-complement activity was qualitatively monitored as the decrease in absorbance of the sample at 410 nm.

The hemolysis assay described above was also used to assess the ability of human recombinant sCR1 to inhibit sheep erythrocyte lysis by complement from other species, such as guinea pigs and rats.
For each species, fresh frozen serum or freshly lyophilized serum or plasma was used as a source of complement. In some cases, serum was obtained commercially (Sigma Chemical
Company, St. Louis, MO).

First, the serum was titrated for its ability to lyse activated red blood cells. The highest dilution that produced at least 80% of the highest erythrocyte lysis was chosen to assess the effect of added human sCR1. Thereafter, human serum was replaced with animal serum and the assay was performed as described above at the preferred dilution.

13.2.2.Results As shown in FIG. 28, purified sCR1 contained 0.12 μg / ml sC
R1 concentration inhibited complement mediated hemolysis by 50% in the main pathway. The ability of antibody affinity purified sCR1 to inhibit the hemolysis assay was compared to the ability of unpurified material (cell culture supernatant containing sCR1). Purified sCR1 has an activity comparable to that of unpurified sCR1, both of which are 50% in a hemolysis assay at 1.6 × 10 ⁸ ghost / ml.
Inhibition. This indicated that the purification procedure did not substantially reduce the functional activity of the final sCR1 product.

Aliquots were stored at -70 ° C for one week to determine if purified sCR1 could be stored frozen. The frozen sCR1 concentration was 28
Same as unfrozen sCR1 as measured by absorbance at 0 nm and CR1 ELISA. Also frozen sC
R1 had the same activity as unfrozen sCR1 as measured by inhibition of hemolysis.

The ability of human recombinant sCR1 to inhibit complement-mediated hemolysis from several species is summarized in Table IX.

Both guinea pig and rat complement appeared to be inhibited by human sCR1. The apparent lack of inhibition for other species is due to (a) the inability to use rabbit antibodies and sheep red blood cells in the assay system;
Or (b) may reflect that high concentrations of serum in this system are required for hemolysis.

13.3. Inhibition of C3a and C5a Production 13.3.1. Methods The ability to inhibit complement was also tested by assaying for specific inhibition of C3a and C5a production. For all experiments, a single pool of human serum to be used as a source of complement was aliquoted and stored frozen at -70 ° C. Human IgG was heat-aggregated and divided into portions, -70
Stored frozen at 0 ° C. For each experiment, a portion of the serum was equilibrated at 37 ° C. with varying concentrations of sCR1 to be tested.
The complement pathway was initiated by the addition of aggregated human IgG. A control sample containing no IgG was always included. After a certain reaction time of 15 minutes (early aging studies have determined that the production of C5a or C3a is almost complete, i.e. determined to have a convenient time interval of more than 90% complete) The level of the complement peptide (C5a or C3a) was determined by a modified procedure using a commercially available radioimmunoassay (called RIA) kit (C
5a, RIA, (Amersham) Cat. No. RPA.520; C3a RIA,
(Amersham, catalog number RPA.518).

Since a competitive immunoassay was used, the complement peptide (C5a
And C3a) concentration changed inversely with counting. The bound number (CB) for the sample was defined as the total count (counts per minute, cpm) measured in the pellet.

The y-axis in FIG. 29 represents fraction inhibition. Fraction inhibition was determined by the CB for the "control without IgG"
-Divided by (CB in “Sample without sCR1”) (“Sample”)
Binding number (CB)-CB in "sample containing sCR1").

13.3.2.Results The activity of purified sCR1 was measured in activated human serum samples
It was assayed by testing its ability to inhibit the production of 3a.

As shown by FIG. 29, under the test conditions, the purified sC
R1 could maximally inhibit C5a production by 100% and C3a production by 60%. 50% inhibition is 5 μg / ml for C5a production and C
It was observed at a sCR1 concentration of 15-20 μg / ml for the production of 3a. These data suggest that recombinant sCR1 inhibits C5 convertase more effectively than C3 convertase.

14. Examples: Demonstration of Functional In Vivo Therapeutic Activity of Soluble CR1 14.1. Soluble CR1 Shows In Vivo Action on Reverse Passive Arthus Reaction The Arthus reaction involves local injection of an antigen, which is then Is a classic immune-induced inflammatory response triggered by reacting with Major biological responses include immune complex deposition, complement fixation, polymorphonuclear (PMN) leukocyte infiltration,
Characterized by lysosomal enzyme release, vasoactive amines, and local tissue damage (Uriuhura, T. and Movat,
HZ, 1966, Exp.Mol.Pathol. 5: 539-558; Cochrane, CG,
1968, Adv. Immunol. 9: 97-162). A modification of the direct Arthus response, the reverse passive Arthus reaction (RPAR), has been used as a model to identify anti-inflammatory drugs (Pflum, L. et al.
R. and Graem, ML, 1979, Agents and Actions, 9, 184-18
9). In RPAR, antibodies are injected locally and antigen is present in the circulation.

When tested in the rat RPAR model, soluble CR1s was able to block the local inflammatory response. The mechanism of action of this soluble CR1 function in vivo is mediated by inhibition of complement pathway enzymes.

14.1.1. Materials and Methods 5 week old female Sprague Dawley rats (CD strain) (Charles River Laboratories), weighing about 100-125 g, Wilming
ton, MA) was anesthetized by intraperitoneal injection of 0.1-0.3 ml of Avertin solution. This solution is a stock solution made with 1 g of tribromoethanol in 15 ml of Amel Amel ethanol.
1: 2 dilution. The back of the animal was shaved. The tail was then warmed first with warm water and then with a heat lamp.
Using a 1 ml syringe, add 0.15 M phosphate buffered saline (PB
S) 5 mg / ml egg albumin (Calbiochem Corp San D
iego, CA) 0.35 ml from the tip of one tail of the tail vein about 2.5-5.
Intravenous injections were made 1 cm (1-2 inches). After 5 minutes, the rats were treated with a 20 mg / ml rabbit Ig fraction of an anti-ovalbumin antibody with an antibody titer of 4 mg / ml (Organon Teknika C
orp, Cappel Division, West Chester, PA) 0.08ml or
20 mg / ml rabbit IgG (Sigma Chemcal Co., ST. Louis, M.
D) 0.08 ml or injected intradermally with PBS. Each injection was performed in duplicate and the area near the injection was marked circularly with a marker pen. The rats were then monitored at 1 hour, 4 hours, and 18 hours. Twenty-four hours later, rats were killed by submersion in dry ice for three minutes. Skin samples were dissected from the injection site. One of the duplicate samples was fixed in 10% formalin for paraffin embedding, and another was frozen for cryostat sections. Tissue sections were prepared and stained with hematoxylin and eosin.

14.1.2. Results Weak RPAR responses (eg, edema and erythema) began to become visible 3-5 hours after intradermal injection of anti-ovalbumin antibody. Reaction size is 3-5mm after 24 hours
The intensity of the reaction gradually increased until
Figure). No reaction was observed in the skin of the nightly rats injected with non-immune rabbit IgG or PBS alone.

Under microscopic examination of tissue sections prepared from the site of the lesion, many acute inflammatory cells were visible, especially in the skin near blood vessels (FIG. 31b). This is typically identified as vasculitis and perivascular inflammation. That tissue is
Extensive infiltration of PMN, presence of red blood cells in connective tissue,
And typical inflammatory symptoms with relaxation of collagen fibers.

14.1.3. Effect of Intradermal Administration of Soluble CR1 A mixture of purified sCR1 was prepared by combining 40μ of 0.75 mg / ml sCR1 with an equal volume of anti-ovalbumin or normal rabbit IgG or PBS. sCR1: anti-ovalbumin mixture or sCR1: rabbit IgG mixture, or sCR1: P
The BS mixture was injected intradermally into rats sensitized intravenously with ovalbumin. Slightly visible lesions appeared at the injection site that received the sCR1 + anti-ovalbumin antibody (No.
30a). As expected, no lesions appeared in the injection sites that received sCR1: rabbit IgG or sCR1: PBS. sC
R1: Microscopic examination of the section of tissue surrounding the anti-ovalbumin injection site showed PMN and clusters of mononuclear cells around the venules, but extensive infiltration of PMN or extravasation of red blood cells. None (Fig. 31a). These data indicate that administration of soluble CR1 resulted in suppression of endothelial cell damage and suppression of the inflammatory response.

To determine the minimum effective amount of sCR1 required to block RPAR in the ovalbumin rat model described above,
0.75mg / mls 10-fold serial dilution of CR1 stock solution (no dilution, 1/1
0,1 / 100,1 / 1,000 and 1 / 10,000) were tested. Each sCR
Add 1 dilution to an equal volume of neat neat or 1/2 dilution anti-
Mixed with ovalbumin antibody. 80μ in each area
Was injected. The ability of sCR1 to inhibit RPAR was dose-dependent, with an effective reduction in edema at 300 ng per site (Table X).

14.2. Pharmacokinetics of sCR1 administered in vivo The biological half-life of sCR1 in vivo was measured as follows. Rats of similar age (6 weeks) and body weight (110-125 g) were injected intravenously with 250 μg of sCR1 in 0.35 ml. At 2, 5, 10, 60 and 24 hours after injection, the rats were sacrificed and blood was obtained from the vena cava. 1-2 ml of serum from each rat was obtained by centrifugation at 1800 rpm for 10 minutes,
The amount of sCR1 in each sample was measured by CR1 ELISA.
Erythrocytes without control rat serum or hemoglobin
A two-fold dilution of purified sCR1 1 μg / ml added in ghost (1.6 × 10 ⁸ ghost / ml) detergent lysate was added to CR1
Used as standard. The results are shown in Table XI.

These data indicate that sCR1 can be detected 24 hours after intravenous injection. At 24 hours, the level of sCR1 in serum was 50% of the peak level observed 10 minutes after injection
Met.

Additional studies in Sprague Dawley rats and cynomorphous monkeys (Table XII) were performed to further characterize the pharmacokinetics of in vivo administered sCR1. Each animal had only ¹²⁵ I-labeled sCR1 (rat was 14-28 × 10 ⁶ c
pm, monkey is 126-153 × 10 ⁶ cpm) or unlabeled (about 1m
g / kg) and a mixture of ¹²⁵ I-labeled sCR1 was injected intravenously. In the 2-431 study, unlabeled sCR1 1 mg / kg or 10
A dose of mg / kg was also tested in the monkey. Blood Table X
Collected from each animal at several time points within the sample time indicated in II. The clearance of sCR1 from blood followed a biphasic pattern in rats and monkeys. The first phase (α) had a short half-life measured in minutes. The 1 mg / kg dose was t _1/2 α = 9.13 min and the 10 mg / kg dose was t _1/2 α = 29 min, and this half-life was dose dependent as shown in the Monkey study. Phase 2 (β)
Showed a longer half-life, measured in hours (Table XII
And FIG. 32). Monkey study results 1.0 and 1
The monkey was shown to have no clinical signs of toxicity observed in monkeys in connection with single intravenous administration at a dose of 0 mg / kg.

14.3. SCR1 reduces infarct size in rats with reperfused myocardial infarction.

In vitro complement pathway as described herein
SCR1, which was able to suppress the activity of C3 / C5 convertase, could also reduce the degree of reperfusion injury in a rat myocardial infarction model in vivo.

Myocardial infarction can be induced in rats by coronary artery ligation. If established within the first few hours after myocardial infarction,
Reperfusion has been shown to reduce infarct size, improve left ventricular function, and reduce mortality
d, E. and Kloner, RA, 1985, J. Clin. Invest. 76: 1713-17.
19). However, reperfusion of severely ischemic but irreversibly injured myocardium can itself cause damage and enlarge. Mechanisms responsible for reperfusion-induced damage can include oxygen-free radicals and damage mediated by cellular calcium overload. Leukocytes, acting alone or in concert with microvascular endothelial cells, can cause this damage. Complement activation may be involved in this process Rossen, RD et al., 1985, Cir. Res.
57, 119-130; Craw-ford, MH et al., 1988, Circulation 7
8: 1449-1458).

14.3.1. Inhibition of complement activation and myocardial reperfusion injury by sCR1 in rats sCR1 was administered to rats that had transient myocardial ischemia and were subsequently reperfused. The mechanism of ischemic but not irreversible damage of reperfused myocardium is associated with a leukocyte-dependent inflammatory response (Romson et al., 1983, Circulation 67: 1016; Mar
tin et al., 1988, Circ. Res. 63: 483; Kraemer & Mullane, 198.
9, J. Pharmacol. Exp. Therap. 251: 620; Litt et al., 1989, Circ.
ulation 80: 1816; Ambrosio et al., 1989, Circulation 80:18
46) This reaction requires complement activation (Maroko
1978, J. Clin. Invest. 61: 661; Crawford et al., 1988, Cir.
culation 78: 1449). Group animals randomly,
Immediately before occlusion of the left coronary artery by suture ligature, a phosphate buffer solution alone (n = 29) or a phosphate buffer solution containing 1 mg of sCR1 (n = 31) was administered by bolus intravenous injection. After 35 minutes, loosen the suture, close the rib cage, return the animal to the gauge,
Seven days later, animals were killed and examined for myocardial infarction (Weisman et al., 1).
988, Circulation 78: 186); the 7-day interval provides a reliable assessment of infarct size and furthermore, sC for infarct healing
Allowed analysis of possible side effects of R1. The aorta of the excised heart from a rat anesthetized with methoxyflurane was cannulated and the coronary artery was perfused initially with Krebs Henseleit solution and then with 30 mM KCl for diastolic arrest. Intracoronary perfusion and
After fixation by immersion in 10% buffered formalin, 2 m of the heart
The m-cross sections were analyzed histologically by calculating the area of normal, total infarction and infarcted left ventricular myocardium within the organ wall. The sum of the fractional areas of all slices was used to calculate myocardial infarction size and the ratio of necrosis in organ wall to total necrosis. All methods and animal care followed institutional guidelines. The survival rates in the group receiving buffer only (24 of 29) and the sCR1 treated group (25 of 31) were similar to death in all but one of the control rats immediately after coronary artery ligation. Ligation of the coronary arteries was judged successful for 22 animals in each group meeting all of the following standards. Histological evidence of rapid electrocardiographic changes, anterior left ventricular wall cyanosis and postmortem myocardial necrosis compatible with ischemia.
The structure was successfully released from all rats except two of the sCR1-treated animals. Analysis of all survivors, including 2 rats that did not achieve reperfusion, showed that sCR1 treatment reduced myocardial infarction size from an average of 16 ± 2% of left ventricular weight in control rats to 9 ± 2% in the sCR1 group (P <0.01)
showed that. The frequency of intramural infarction was also lower in sCR1 treated (6 of 25) than in control rats (12 of 24) (P <0.0
Four).

The thickness of the infarct segment summed from 4 sections of heart from all sCR1-treated rats was 7.8 ± 0.4 mm, which was not significantly different from that of all untreated animals, 7.3 ± 0.4 mm. Slightly less than that of the non-infarcted intraventricular septum far from the hearts of these two groups (sCR1 rats 9.3 ± 0.2 mm, untreated rats 9.4 ± 0.2%).
0.2mm). In addition, there was no difference in ventricular cavity size between these two groups (sCR1 rat 64.9 ± 3.6 mm ³ , untreated rat 68.9 ± 2.6 mm ³ ). Thus, judging from cardiac observation one week after infarction, sCR1 suppresses myocardial infarction size, but does not cause ventricular dilation or thin the ventricular wall to prevent healing.

Another group of buffer-treated (n = 7) and sCR1-treated rats (n = 8) was used to determine whether suppression of tissue damage by sCR1 was associated with inhibition of complement activation by ischemic myocardium. Animals were sacrificed 3 hours after reperfusion following the same ischemia reperfusion protocol. Nitroblue tetrazolium (NBT) staining (Lillie, RD, 1965, Hist) to detail the area of irreversible injury from the surviving myocardium to the heart
opathologic Technic and Practical Histochemistry, M
cGraw-Hill, New York ed: 3: 378) and rat C5b.
-9 membrane attack complex (Schulze et al., 1989, Kidney Int. 35: 6
Immunoperoxidase staining with a mouse monoclonal antibody against 0) (DeLellis, in Basic Techniques o
f Immunohistochemistry, DeLellis, Ed., Masson, New Yor
k, 1981). The mouse peroxidase-antiperoxidase (PAP) system was used for immunostaining as described. 0.05M for every step of the method
Three 10 minute washes in Tris buffered saline were performed first. Sections were fixed with acetone and 5% with 0.5% H ₂ O ₂ -methanol solution.
For 1 minute with 4% heat-inactivated goat serum. Subsequently, they were treated with a 2 μg / ml primary mouse monoclonal antibody for 18 hours using a mouse antibody (Organon-Tecknika, West C.
affinity-purified F (ab ') ₂ goat antiserum to Hester, PA) for 60 minutes and mouse PAP for 60 minutes, sequentially incubated at room temperature. Slides were developed with 3,3'-diaminobenzidine tetrahydrochloride and counterstained with Gill's hematoxylin 3. In NBT-negative, infarcted areas of control rats (n = 7), C5b-9 complex was mainly dependent on the endothelium of capillaries and small arteries but not on myocardial fibers. In contrast, in rats receiving sCR1 as exemplified in the representative sections shown, the NBT-negative regions were consistently reduced in size, and C5b-9 complex was little or no in these regions. Could not be detected.

Quantification of leukocytes in these serial sections (Ambros
io, et al., 1989, Circulation 80: 1846) reveal that in the infarct zone of control rats, capillaries and venules almost exclusively have 195 ± 28 leukocytes per mm ² (150 × field; n = 3). became. The corresponding section from the heart of a rat receiving sCR1 contains only 83 ± 3 leukocytes per mm ² (n = 4; P = 0.006), and inhibition of complement activity is associated with reduced accumulation of this inflammatory cell type. Along the endothelial surface shown to be
Localization of the C5b-9 complex indicates that these cells are a major part of complement activation in the pathogenesis of re-infusion damage to ischemic myocardium, and the infarcted myocardium 24 hours after coronary artery closure. In contrast to the more diffuse and scattered complement proteins throughout (McManus et al., 1983, Lab. Invest. 48:43
6). The latter merely reflects the ability of necrotic tissue to activate complement, while the former indicates that ischemic stressed endothelium gains complement activation function. Complement activation by endothelial cells is associated with intravascular leukocyte activity of C5a and CR1
Would result in their rapid upregulation of cell receptors, including CR3, and would be a particularly powerful stimulus for early localization of neutrophils to ischemic myocardium (Fearon
& Collins, 1983, J. Immunol. 130: 370; Arnaout et al., 1984,
J. Clin. Invest. 74: 1291). The latter receptor is CR3, iC3
b has been shown to promote the attachment of neutrophils to complement activated endothelial cells bearing the ligand of b (Marks et al., 1989, Na
ture 339: 314). Thus, inhibition of complement activation by sCR1 clearly explains the reduced number of neutrophils attached to endothelial cells.

Reperfusion of ischemic myocardium with thrombolytic agents reduces infarct size, improves left ventricular function, and reduces mortality if established within hours of coronary atresia (Guerci et al., 1992).
87, N. Engl. J. Med. 317-1613; ISIS-2 Collaborative Gr
oup, 1988, Lancet ii: 49; Van de Werf & Arnold, 1988, B
r.Med.J.279: 1374). However, the potential benefits of reperfusion may not be fully achieved. This is because re-infusion into extremely ischemic but irreversibly undamaged myocardium induces necrosis (Becker & Ambrosio, 1).
987, Prog. Cardiovasc. Dis. 30:23; Braunwald & Kloner,
1985, J. Clin. Invest. 76: 713). Two things that may be causally linked to necrosis are neutrophil intravascular accumulation and microvascular endothelial cell damage (VanBenthuysen et al., 1987, J. Med.
Clin. Invest. 79: 265), both are the result of complement activation. This finding of sCR1 cardioprotection supports a possible central role for complement, extends early studies where complement is depleted from cobra venom factor, and enhances the potential for helping thrombolytic tissue Provide a means to do so.

14.3.2. Conclusions The results show that sCR1 treatment is effective in reducing reperfusion in vivo and improving the effects of myocardial infarction. To the extent reperfusion injury is ameliorated, the absolute amount of rescued myocardium increases and the time period during which reperfusion is clinically useful is extended. Treatment with sCR1 is described in the next section or is a useful co-therapy with thrombolysates for pulmonary coronary angiogenesis during acute infarction.

15. Examples Soluble complement receptor 1 (sCR-1) and p-
Co-formulation of anisoylated human plasminogen-streptokinase-activator complex (APSAC) Purified sCR-1 [0.93 mg in sterile dalbecophosphate-buffered saline, prepared as described in Section 12.2.
0 ml] was added to a vial of APSAC reconstituted in sterile water (4.0 ml). The APSAC preparation contained the following:

APSAC: 30 units D-mannitol 100 mg human serum albumin EP 30 mg p-amidinophenyl p'-anisate (anisat
e) HCl 0.15 mg L-Lysine HCl 35 mg 6- amino hexanoic acid 1.4 mg (all figures are subject to normal analytical variation) solution was mixed well and the vial with solid CO ₂ -78
Frozen to ° C. The product was lyophilized at 2-3 mbar / -60 ° C (compressor temperature) for 24 hours and the vial was stoppered again and placed at -70 ° C. To restore, transfer vial to 0.1 M Hepes, 0.15
Dissolved in M NaCl pH 7.4 (1.0 ml) and placed on ice.
Assay dilutions were made in this Hepes buffer. As a control, a vial of APSAC only (same batch) was restored in the same manner, and a freshly opened sample of sCR-1 (same batch) was also tested. The samples were assayed for inhibition of complement-mediated hemolysis according to the method described in Section 13.2.1, and the results are shown in Table XIII below.

Curves fitted to data in Table XII inhibit hemolysis by 50%
The values of the concentration of CR-1 are shown, which are 15.7 ± 3.0 mg / ml and 16.0 ± 2.9 ng / ml for sCR-1 / APSAC and SCR-1 alone. These figures are not different and the results indicate that the activity of sCR-1 is not affected by APSAC medical dosage forms and co-formulation.

16. EXAMPLES Molecular definition of the F 'allotype of human CR1, C3b
Loss of binding sites is associated with a functional change.

Human CR1 consists of a tandem long homologous repeat (LHR) segment that encodes separate binding sites for C3b or C4b. Homologous recombination with unequal crossovers has been proposed as a genetic mechanism that results in a CR1 allele that differs in their total number of LHRs. F allotype has 4LHR,
From 5 'to 3', they are named LHR-A, -B, -C, -D. The site in LHR-A binds preferentially to C4b and LHR-B
And those of -C prefer C3b. Previous studies have revealed the presence of a fifth LHR with a similar sequence to LHR-B and a third C3b binding site in the high molecular weight S allotype. In this study, an 18 kb EcoR V fragment that is associated with the expression of low molecular weight F 'allotypes hybridized to the unique pattern of cDNA and LHR-C specific intron probes. Deletion of LHR-B and one C3b binding site has been proposed as the mechanism by which this F'-specific fragment appears. The molecular mechanism for the association of F 'allotypes with SLE was explored by functional comparison of soluble sCR1 with 1, 2, or 3, C3b binding sites. These three mutants did not show any significant differences in their ability to act as cofactors in the cleavage of monomer C3b, but their relative binding of dimer ligands changed more than 100-fold. Furthermore, variants with only one C3b binding site were at least 10-fold more effective at inhibiting the alternative pathway C3 and C5 convertases. These observations indicate that the F 'allotype interferes with its ability to bind opsonized immune complexes, inhibit the formation of alternative pathway C3 and C5 convertases, and possibly mediate other CR1-dependent cellular responses. Is suggested. They also show that some functions of the CR1 molecule are enhanced by increasing the degree of binding of the C3b binding site.

16.1. Introduction Describes four allotypic forms of CR1 that differ in size by a 30-50 KD increase. The transcripts associated with each allotype also differ by -1.4 Kb increments, indicating that their primary sequence varies in the number of LHRs (Wong et al., 1983, J. Clin. Invest. .72: 685; Dykman et al., 198
3, Proc. Natl. Acad. Sci. USA 80: 1698; Dykman et al., 1984,
J. Exp.Med. 159: 691; Dykman et al., 1985, J. Immunol. 134: 178.
7; Wong et al., 1986, J. Exp.Med. 164: 1531; Holers et al., 1987, P.
roc.Natl.Acad.Sci.USA84: 2459). The F (or A) allotype of ~ 250 KD has 4 LHRs, from 5 'to 3' LHR-A, -B, -C, and -D (Examples 6 and 7,
Supra; Wong et al., 1989, J. Exp. Med. 169: 847). The first two SCRs of LHR-A determine its ability to bind C4b, while the corresponding components of LHR-B and -C determine their high affinity for C3b (see Example 9, Genome, supra). Larger S (or B) of ~ 290 Kd according to restriction map of phage clones
Analysis of the gene encoding the allotype revealed that the 5 ′ half was LHR-B and the 3 ′ half was a chimera of LHR-A,
A fifth LHR, predicted to contain a C3b binding site, was identified (Wong et al., 1989, J. Exp. Med. 169: 847). SLE
The minimum F '(or C) allotype of ~ 210 KD CR1 is associated with an increased incidence in patients with multiple lupus homes (Dykman et al., 1984, J. Exp. Med. 159: 691; Van D
yne et al., 1987, Clin. Exp. Immunol. 68: 570) resulting from the deletion of 1 LHR, which interferes with its ability to bind efficiently to immune complexes that are coated with complement fragments. The data presented below use an intron probe specifically for each LHR to define the molecular basis of the F 'allele and to identify the EcoR of CR1 present in individuals expressing this allotype.
Analyze RV restriction fragment length polymorphism (RFLP). To provide an explanation for the apparent association of F 'allotypes with SLE, a soluble sCR1 mutation with 1,2, or 3, C3b recognition sites and corresponding to the predicted structure of each F', F, S allotype Species were compared for their ability to bind dimer ligands, act as cofactors in the cleavage of C3b, and inhibit alternative and major pathway C3 and C5 convertases.

16.2. Materials and Methods Analysis of DNA Genomic DNA was prepared from peripheral blood leukocytes, digested with EcoR V, electrophoresed and analyzed by Southern blot as previously described (Wong et al., 1989, J. Exp. Med. 164: 153
1). The CR1 cDNA probe 1-1 hybridizes to all SCR-4 to -7 of the LHR, whereas the probe 1-4 is LHR-
Specifically hybridizes to SCR-1 and -2 of B and -C (Klickstein et al., 1987, J. Exp. Med. 165: 1095;
Klickstein et al., 1988, J. Exp.Med. 168: 1699, Wong et al., 198
9, J. Exp. Med. 169: 847). Non-cord probe PE is LHR-
Hybridizes with the intron between the two exons encoding SCR-2 of B and -C. PX is LHR-A and-
HE hybridizes to the intron between the two exons encoding SCR-6 of B, and HE hybridizes to the 5 'intron of SCR-7 of LHR-A and -B (Wong et al., 198).
9, J. Exp. Med. 169: 847) (Fig. 34). CR1 of a genomic library derived from the DNA of an individual whose F allele is homozygous
Clones were screened by hybridization to a cDNA probe spanning the entire coding sequence. Restriction mapping of overlapping phage clones spanning the F allele was performed as previously described (Wong et al., 1989, J. Exp. Med. 169: 847).

Construction of Expression Plasmid and Purification of Soluble sCR1 A Pst I fragment extending from the SCR-R of LHR-A to the corresponding position of LHR-B was isolated from plasmid pBSABCD containing the complete coding sequence of the F allotype of CR1 (see above). Example 8.1). This was inserted into piABCD (described above, Example 8.2) linearized by partial digestion of Pst I to create a fifth LHR having the same coding sequence as that of LHR-B.
A clone named piABBBCD having an in-frame insert was selected from a restriction map, the portion encoding the entire extracellular domain was cut by digestion with XhoI and ApaLI and treated with Klenow DNA polymerase. Ligating the blunt-ended fragment with the linker 5'-TGAGCTAGCTCA-3 ',
Removed with Nhe I, CDM8-derived product, Xba I of expression vector Ap ^r M8
It was inserted into the site (Seed, 1987, Nature 329: 840). pasecA
This plasmid, designated BBCD, has a stop codon inserted after the 37th obtained SCR and lacks the transmembrane and the sequence encoding the cytoplasmic domain. Plasmid pasecABCD was created by transferring 4 LHR of the F allotype from pBSABCD to Ap ^r M8 using a similar method. A third plasmid, containing only 3 LHR and designated pasecACD, was created by in-frame deletion of the Pst I fragment spanning from SCR-5 of LHR-A to the corresponding position of LHR-B. 30 for each of the above plasmids
-60 μg of high glucose and 10% Nuserum (Collabora
400 μg in Dulbecco's modified Eagle's medium (DMEM) (Hazelton, Lenexa, KS) with tive Research, Bedford, MA).
2 x 10 ⁷ COS-1 cells (American Typ.
e Culture Collection, Rockville, MD). After removing the transfection medium, the cells were incubated at room temperature in HBSS without divalent cations.
% DMSO for 3 minutes (Ausubel et al., 1987, in C
urrent Protocols in Molecular Biology, John Wiley
& Sons and Greene Assoc., New York), washed and cultured in DMEM and 10% FCS. The culture supernatant was collected every 48 hours for 10 days, and cell fragments were removed by centrifugation and frozen at -70 ° C. The supernatant was thawed and PMSF and sodium azide were added to a final concentration of 5 mM and 0.2%, respectively, and the detergent was omitted from the elution buffer (Wong et al., 19
85, J. Immunol. Meth. 82: 303; Example 12.2.1, supra ). Purified by mAb YZ-1-Sepharose by affinity chromatography as described. The purified protein was dialyzed twice against 1000 volumes of PBS and frozen in small aliquots at -70 ° C. This method usually yielded 150-200 μg of sCR1 as determined by the Micro BCA kit using BSA as a standard (Pierce, Rockford, Ill.).
Proteins were analyzed by SDS-PAGE on a gel containing a 5% -15% acrylamide linear gradient.

Cofactor activity of sCR1 Purified human C3 (Tack & Prahl, 19
76, Biocehmistry 15: 4512) at 37 ° C for 5 minutes with 0.5% TPCK-
Treated with trypsin (Sigma, St. Louis, MO) and quenched by adding a 4-fold molar excess of soybean trypsin inhibitor. 5 using Iodogen (Pierce, Rockford, IL)
C3b was labeled with ¹²⁵ I until it had a specific activity of × 10 ⁵ . C3B
200 ng, Insi I 100 ng (Fearon, 1977, J. Immunol. 119: 124
8) was incubated with various amounts of sCR1 in 20 ul PBS for 1 hour at 37 ° C. to determine the cofactor activity of sCR1 (Wong et al., 1985, J. Immunol. Meth. 82: 303; Example 11, the above). SDS containing 0.1 M dithiothreitol
The reaction was stopped by boiling the sample in an equal volume of PAGE sample buffer. After electrophoresis and autoradiography, the area of the dried gel corresponding to the position of the α 'chain was excised,
The amount of radioactivity was measured with a Beckman gamma counter (W
ong et al., 1985, J. Immunol. Meth. 82: 303). The count associated with the α 'chain in the absence of CR1 was taken as a 100% control.

The capacity of sCR1 to bind dimer C3b C3b was converted to dimethisperimidate (Sigma, St. Louis, MO) (Wilson et al., 1982, New
Engl. J. Med. 307: 981) or 1,6-Bismaleimidohexane (Pierce, Rockford, IL) (Weisman et al., 1990, Scienc).
en 249: 146) and dimers were selected by sedimentation with a primary gradient of 4.5% to 30% sucrose in PBS (Wi
lson et al., 1982, New Engl. J. Med. 307: 981). Either method has an erythrocyte CR with an association constant (Ka) in the range of 1-3 × 10 ⁸ M ^−1.
A dimer linked to 1 was obtained. 300 ng of ¹²⁵ I-C3b dimer
(4 × 10 ⁶ cpm / ug) in 200 ul of HBSS containing 0.1% BSA in the presence or absence of increasing amounts of unlabeled monomer or dimer C3b or different soluble forms of sCR1 , 2 × 1
0 incubated with ⁸ red blood cells (Weisman et al., 1
990,249: 146). After 1 hour on ice, cell-bound ligand was separated from unbound material by centrifuging erythrocytes through dibutyl phthalate (Wilson) et al., 1982, N.
ew Engl. J. Med. 307: 981). The amount of dimer C3b bound in the presence of excess rabbit IgG anti-CR1 was taken as the non-specific background value, and the specific count bound in the absence of any inhibitor was taken as the 100% control. For all these binding tests, red blood cells from one normal individual were used. These cells were homozygous for the F allotype and had a relatively large amount of CR-1 (-800YZ-1 mAB binding site).

Inhibition of alternative and conventional pathway convertase Veronal with ₂ mM MgCl2 and 8 mM EGTA in the absence or presence of increasing amounts of sCR1 to assess alternative pathway activation
5 × 10 ⁶ thymosan particles (Dr. Joyce Czop, H.
arvard Medical School, Boston, MA) with 25% human serum. To assess the activation of the conventional pathway, heat aggregated rabbit IgG 60 ug / ml was replaced with thymosan and Veronal with 0.5 mM MgCl ₂ and 0.15 mM CaCl _3.
Reactions were performed in buffered saline (Weisman et al., 1990, Scien).
ce 249: 146). After incubation at 37 ° C for 40 minutes, 10
The reaction was stopped by adding mM EDTA, and the reaction was stopped using a radioinomic assay kit (Amersham, Chicago, IL).
The amount of decomposition of 3a and C5a was measured.

16.3. Results Structure of CR1 F 'Allele When the DNA of an individual expressing the -210 kD F' allotype of CR1 was digested with EcoRV, CR1
It was previously reported that a further 18 kb fragment was observed with the Southern blot probe using cDNA probe 1-1 (Wong et al., 1986, J. Exp. Med. 164: 1531) (FIG. 33). Although this probe was originally derived from SCR-3 to SCR-7 in LHR-B, its sequence had sufficient homology to hybridize to the fourth to seventh positions of other LHRs ( Examples 6 and 7, supra; Klickstein et al., 1987, J. Exp. Med. 165:
1095; Klickstein et al., 1988, J. Exp.Med. 168: 1699; Wong
Et al., 1989, J. Exp. Med. 169: 847). To assign each fragment to an LHR, the EcoR V map of overlapping genomic clones across the F allele was examined (FIG. 34). 9.4kb and 22k
The fragment of b corresponded to that deduced from LHR-A and -D, respectively. This goes to intron probes PX and HE
It was confirmed that the 9.4 kb fragment could hybridize but the 22 kb fragment could not (FIG. 33). EcoR V for LHR-B
Since no sites were present, the largest fragment at the top of the plot hybridized to all probes showed a -32 kb fragment spanning LHR-B and most LHR-C (Figures 33 and 3).
Four). A previously reported pseudogene similar to CR1 (Wowg et al., 1989, JE
xp.Med.169: 874), the restriction maps of the genomic clones indicate that the 20 kb and 15 kb fragments are derived from this region (No. 33).

The 18kb EcoR V fragment involved in the expression of the F 'allotype is cDN
It hybridized to A probe 1-4 and intron probe PE and was shown to contain the 5 'half of LHR-B and -C. This fragment is an intron probe
LHR because it did not hybridize to PX and HE
It was shown that the 3 'half of -B was deleted (Fig. 3
3). Deletion of the LHR-B or another fragment of similar length from the -32 kb EcoR V fragment will result in the most 3 'EcoR V of LHR-A.
18k extending from the site to the 5'-most EcoR V site of LHR-C
This results in a fragment of b (FIG. 34). Such fragments were hybridized only to probes 1-4, PE and 1-1,
Consistent with 33 results. The deletion of any -15 kb, including the PX and HE sites of LHR-B, necessarily includes at least one exon, including the C3b binding site of either LHR-B or -C, resulting in three LHRs. Allele encodes only one C3B binding site.

Plasmid Construction and Purification of Soluble sCR1 A direct comparison of the affinity of the CR1 allotype for opsonized immune complex or multimeric ligands is due to the absence of individuals homozygous for the F 'allotype. Not done. Using the sequence of complete CR1 and cDNA and the results of genomic analysis, sCR1 with the predicted structure for various polymorphic variants could be generated. Insert a stop codon between the end of the 3'-most SCR and the start of the transmembrane domain to evaluate the capacity of individual CR1 molecules of each allotype for divalent interactions with C3b dimers Thus, cDNA for soluble CR1 was constructed. By this method, the dimer C3
Possible interactions between b and the distantly adjacent molecule of membrane-bound CR1 can be ruled out. Furthermore, by using soluble CR1,
The potential for interference in enzyme assays with detergents necessary to maintain the solubility of the membrane-derived preparation can be eliminated. The methods used for insertion or deletion of LHR utilize conserved restriction enzyme sites to preserve reading frames. The inserted or deleted Pst I fragment is LHR-
SCR-5, -6, -7 of A and SCR-1, -2, -3 of LHR-B
And -4 (FIG. 35). Since the amino acid sequences of the third to seventh SCRs are identical in LHR-A and -B (FIG. 35) (Examples 6 and 7, above), insertion or deletion of a sequence equivalent to LHR-B by these methods is performed. Loss is made. Therefore, plasmids pasecACD, pasecABCE,
And pasecABBCD (Figure 35) contain one, two or three C3s
b Encodes proteins that each have a binding site.

Soluble recombinant CR1 was isolated from the culture supernatant of COS cells transfected with the CR1 plasmid by chromatography on YZ-1-Sepharose. Each sCR1
The protein is more than 95% pure and the three forms are
Similar to that observed in the natural allotype, SDS-PAG
E showed a progressively increasing Mr difference at −30 kb under non-reducing conditions (FIG. 35). The biosynthetic labeling with ²⁶ S- cysteine COS cells transfected with the vector alone Ap ^r M8,
It did not show absorption of CR1-like protein into YZ-1 Sepharose. In addition, soluble sCR1 has a similar Mr to CR1 isolated from ¹²⁵ I-labeled erythrocytes, which is only 7% from each molecule.
This was consistent with the deletion of 0 amino acids. pasec
In lanes containing sCR1 isolated from cells transfected with ABBCD- or pasecABCD, small amounts of proteins with similar Mr to the smaller form were observed (lanes 2 and 3). These represent homologous recombination products naturally occurring in transfected COS cells.

Cofactor activity of sCR1 radiolabeled C3b to iC3b and C3dg
Various forms of sCR1 in cofactor assay converted to fragments
Functional integrity was measured. 50% factor I of the α 'chain of C3b
The amount of sCR1 required for the mediation field is 3.
From 5 nM to 8 nM for pasecACD-derived protein, the difference between the three forms was slight (FIG. 37). Furthermore, the conversion from C3b to C3dg is 1
This was observed with the addition of all forms of sCR1 from 0 nM to 20 nM. That is, the soluble recombinant CR1 retains the capacity of a natural molecule that binds to C3b, regardless of the number of C3bc binding sites,
Acted as a cofactor for factor I degradation.

Capacity of sCR1 binding to dimer C3b To assess the effect of having different numbers of LHR-B on the binding of C3b coated targets, the ¹²⁵ I-C3b dimer by erythrocyte CR1 was evaluated using sCR1 mutant Inhibition of uptake. The concentration required for 50% inhibition reflected the relative binding of either the ligand or the cell-bound receptor. In the experiment shown in FIG. 38, unlabeled C3b dimer 10 nM was compared with ¹²⁵ I-C3b dimer and red blood cells.
Required for 50% inhibition of the interaction with CR1. The interaction of the receptor with monomeric C3b was rather weak, requiring 1 uM or more than 100-fold of this ligand to achieve similar inhibition (FIG. 38). The low binding of this monomeric interaction is such that crosslinking of two different molecules of soluble CR1 by a dimeric C3b ligand is undesirable under these conditions,
It shows that occupancy of two intramolecular binding sites is required for effective antagonism. This bivalent interaction was confirmed by the requirement of 10 nM and 100 nM of sCR1 from pasecABCD and pasecACD for 50% inhibition, respectively. Interestingly, pas with three C3b binding sites
Only 1 nM of sCR1 from ecABBCD was required for a similar effect (FIG. 38). That is, the soluble CR1 forms having one, two or three C3b binding sites differed in their dimeric C3b binding properties, but the monomeric forms of this ligand used as substrates for factor I degradation Was not different (Figure 37).

Inhibition of alternative and conventional pathway-adding enzymes Compare the volume of soluble CR1 mutants that block alternative and conventional pathway-adding enzymes by measuring differentially released C3a and C5a upon incubation of human serum with zymosan or aggregated IgG did. Only 1-2 nM pasecAB to achieve 50% inhibition of both alternative pathway C3 and C5 convertases
Whereas sCR1 from BCD or pasecABCD was required, sCR1 of pasecACD required more than 30-fold concentration to achieve a similar effect (FIGS. 39A and 40A). This was consistent with the above prediction that the sCR1 mutant binds differently to the C3b homodimer in C5 convertase (Kinosh
Ita et al., 1988, J. Immunol. 141: 3895). A similar effect was observed at the C3 point or the enzyme, indicating that properdin stabilized C3
This indicated the presence of multiple C3b molecules involved in the bBb complex. In contrast, the conventional pathways C3 and sCR1
No difference was observed in the capacity to inhibit C5 convertase (FIG. 39).
B and 40B), indicating that they also bound to the C4bC2a complex of C5 convertase or to the C4b molecule in the C4b / C3b heterodimer (Takata et al., 1987, J. Exp. Med. 165: 149).
Four). In addition, except for the pasecACD-derived molecule, inhibition of the enzymes of the conventional pathway compared to the alternative pathway requires 2-10 fold higher concentrations of recombinant CR1, which is probably due to C4b
This was thought to reflect the low binding of sCR1 to the contained invertase.

16.4. Discussion Each polymorphic variant of human CR1 is encoded by a different number of LHRs, as predicted by a -1.3 kb difference in transcription associated with each allotype (Wong et al., 1986). , J.
Exp.Med.164: 1531; Holers et al., 1987, Proc.Natl.Acad.Sc.
i.USA.84: 2459). The parallelism between the homology of the different LHR coding regions of the CR1 gene and the homology of the corresponding non-coding regions has made it possible to predict the coding sequence based on the restriction map of the genomic clone. That is, LHR−
The fifth LHR in the S allele with the 5 'half similar to B and the 3' half similar to LHR-A was predicted to encode a third binding site for C3b within the S allotype ( Wong et al., 1989, J. Exp. Med. 169: 847). Other 19
EcoR V is the only restriction enzyme that has revealed the RFLP associated with the F 'allotype (Wong et al., 1989, 1989).
J. Exp. Med. 164: 1531), a deletion in this allele apparently involved a region of high homology. In this test, an 18 kb EcoR V fragment was involved in individuals expressing the F ′ allotype hybridized specifically with a combination of probes derived from exons and introns to LHR-C (Wong et al., 1989). , J.Exp.Med.164: 153
1). This pattern was consistent only with a deletion of LHR-B or an equivalent region that resulted in the loss of one C3b binding site. The appearance of the 18 kb fragment was accompanied by the disappearance of the -30 kb fragment from LHR-B or -C on Southern blots, and individuals homozygous for the F 'allele could not be obtained for comparison. Could not be confirmed. Smaller transcripts for this allotype are thought to appear through alternative splicing (Wong et al., 198
6, J. Exp. Med. 164: 1531; Holers et al., 1987, Proc. Natl. Aca.
d.Sci.USA.84: 2459), these mechanisms are similar to the observed EcoR
V Cannot explain RFLP.

The involvement of the F 'allotype with SLE8 is consistent with the suggestion that the capacity of this variant to bind to opsonized immune complexes is reduced (Dykman et al., 1984, 1984).
J. Exp. Med. 159: 691; Van Dyne et al., 1987, Clin. Exp.
l.68: 570). The functional group capacities of sCR1 with different numbers of LHR-B were compared. Effective antagonism of C3b dimer uptake by soluble CR1 in a recent study indicates that occupancy of the tandem binding site occurs (Weisman et al., 1990, Science 249: 146), i.e. It became possible to directly evaluate the binding of each receptor molecule. Our method was purified from culture supernatant of COS cells transiently transfected with a plasmid encoding a CR1 molecule with or without the inserted stop codon and one or two LHR-Bs. In both the use of soluble CR1,
It is different (Figure 35). Monomers or dimers required to inhibit binding of radiolabeled dimer C3b to erythrocytes by 50%
The amounts of recombinant CR1 from C3b and pasecABCD were very close to those previously reported (Weisman et al., 1).
990, Science249: 146). Further purification on dimer C3b
CR1 binding was increased 10-fold with the addition of each C3b binding site, producing a 100-fold difference between sCR1 molecules derived from pasecACD or pasecABBCD (FIG. 38). Consistent with previous observations that the C4b binding site within LHR-A retains low affinity for C3b (Klickstein et al., 198
8, J. Exp. Med. 168: 1699), pasecACD-derived sCR1 with only LHR-A and -C was more effective than C3b monomer in blocking the uptake of radiolabeled dimer ( (Figure 38). The higher binding observed for pasecABBCD-derived CR1 compared to pasecABCD-derived molecules also suggests expansion of the two pairs of binding sites (FIG. 38), It confirmed the previous hypothesis that linear interactions are preferred for multivalent interactions (Klic
Kstein et al., 1988, J. Exp.Med. 168: 1699; Wong et al., 1989, JE.
xp.Med.169: 847). Due to the relatively low binding of CR1 to monomeric C3b (FIG. 38), various s in the present study (FIG. 37) and other studies (Seya et al., 1985, J. Immunol. 135: 2666).
The absence of an apparent difference in the cofactor capacity of the CR1 form indicates that the conditions used were not suitable for co-occupation of more than one active site by the monomeric ligand.

Soluble blocking alternative enzymes C3 and C5 convertase
Differences in the efficacy of CR1 mutants are consistent with their different affinities for C3b homodimers (Kinoshita
Et al., 1988, J. Immunol. 141: 3895). In fact, in this assay, sCR1 mutants with two or more C3b binding sites
It was at least 30-fold more effective than one with only one site (FIGS. 39A and 40A). However, pa
Although the binding of secABBCD-derived sCR1 was significantly higher in liquid phase assays (FIG. 38), it was not as effective as that of pasecABCD-derived (FIGS. 39A and 40A). That is, it is believed that the maximum volume in this assay is limited by the topographical distribution of invertase sites on the activated surface. All of these variants tested retained one site for C4b and at least one adjacent site for C3b,
The structure is consistent with their comparable capacity to inhibit the invertase of the conventional pathway (FIGS. 39B and 40).
B) (Takata et al., 1987, J. Exp. Med. 165: 1494). The difference between our observations and the results of another study is that F (or A) and F '(or C) were used instead of the less stable liquid phase invertase and instead of the purified form. (Seya et al., 1985, J. Immunol. 1)
35: 2661). It is also explained by the fact that the F 'mutants used in the test had different compositions of LHR.

The efficiency of uptake of immune complexes by CR1 is enhanced by population formation of this receptor on erythrocytes (Paccaud et al.
988, J. Immunol. 141: 3889; Chevalier & Kazatchkine, 19
88, J. Immunol. 142: 2031), the population of this receptor in individuals with low numbers of CR1 is smaller or the number of receptor molecules per population is smaller (Paccaud et al., 1988, J. Immunol.
141: 3889). Thus, the presence of F 'allotypes and small amounts of CR1 in SLE patients results in a reduced total amount of C3b and C4b binding sites and a reduced efficiency of immune complex clearance from the circulation (Dykman et al., 1984, J. .Exp.Med.15
9: 691; Van Dyne et al., 1987, Clin.Exp.Immunol. 68: 570; Wil.
Son et al., 1987, Immunol. Res. 6: 192; Miyakawa et al., 1981, Lan.
cet ii: 493; Schifferli et al., 1988, J. Immunol. 140: 89.
9). The low affinity of sCR1 observed with conventional pathway invertases (FIGS. 39 and 40) indicates that the relative amounts of C4b and C3b deposited on soluble immune complexes are important for CR1-dependent uptake and processing ing. Absence of pre-populated CR1 in neutrophils under some conditions (Paccaud et al., 1990, Eur. J. Immunol. 20: 283).
Suggest that the mechanism that triggers receptor aggregation is important for initiating a biological response in nucleated cells (Daha et al., 1984, J. Immunol. 132: 1197; Bacle et al.) , 1990,
J. Immunol. 144: 147). It is believed that shorter CR1 allotypes further reduce the efficiency of such interactions and result in impaired receptor mediator cell responses at sites of tissue inflammation.

17. Deposit of microorganism Escherichia coli DK carrying plastic piABCD encoding full-length CR1 protein (referred to as pCR1-piABCD)
1 / P3 at Agricultural Researc in Peoria, Illinois
Deposited with the h Culture Collection (NRRL) on March 31, 1988, and assigned the registration number B-18355.

Plasmid pBSCR1c / pTCSg encoding soluble CR1 molecule
Chinese hamster ovary cell line DUX B11 carrying pt clone 35.6 was transferred to Am, Rockville, Md.
March 1989 in the erican Type Culture Collection (ATCC)
Deposited on the 23rd and assigned registration number CRL10052.

The present invention should not be limited in scope by the deposited microorganism. The deposited embodiment is intended as merely an example of one feature of the invention, as any microorganism that is functionally equivalent is within the scope of the invention. Indeed, various modifications of the invention in addition to those shown and described herein will become apparent to those skilled in the art from the foregoing description and accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

It is also understood that all base pair sizes given for nucleotides are approximate and are used for illustrative purposes.

Various documents are cited herein, the disclosures of which are incorporated by reference in their entirety.

──────────────────────────────────────────────────続 き Continued on the front page (73) Patent holder 999999999 Avant Immunotherapeutics, Inc. United States Massachusetts 02139, Cambridge, Sydney Street 38 (72) Inventor Fairron, Douglas, Ty. Maryland, United States 21210 Britain Wood Road, Baltimore, 2nd (72) Inventor Clickstein, Lloyd, Be. 02146 Massachusetts, USA, Brooklyn, Apartment 8, St. Paul Street 32 (72) Inventor Wong, Winnie, AW. 02159, Massachusetts, USA Newton, Parker Street 9 (72) Inventor Carson, Gerald, A U.S.A., Massachusetts 02181, Welsley, Welsley College, Ho McCarpie (no address) (72) Inventor Concino, Michael, F. United States Massachusetts 02159, Newton, Asselstein Road 76 (72) Inventor Ip Sadbury, Massachusetts, United States 01776, Singbury, Singing Hill Circle 11 (72) Inventor McClids, Sabbath, Sea. United States 01730, Massachusetts, United States 01730, Bedford, Fletcher Road 37 (72) Inventor Marsh, Henry, C. Jr., Jr., USA 01867, Massachusetts, Reading, High Street 218 (56) References. Exp. Med. , Vol. 168, p. 1255-1270 (1988) (58) Fields investigated (Int. Cl. ⁷ , DB name) C12N 15/00 A61K 38/00 A61P 7/02 A61P 9/10 C07K 14/00 BIOSIS (DIALOG) WPI (DIALOG)

Claims (14)

(57) [Claims] (1) A soluble complement receptor type 1 (sCR1) molecule or a full-length complement receptor 1 substantially comprising the amino acid sequence shown in FIG. 1 and lacking a transmembrane region.
A fragment thereof retaining the biological activity of the type molecule, and (2)
A composition comprising a thrombolytic agent in an amount effective for use in combination with a method of treating a thrombotic condition in a human or other animal, respectively. 2. The soluble complement receptor type 1 molecule or a fragment thereof comprises: (a) C3b binding; (b) C4b binding; (c) C3b binding and C4b binding; (d) Factor I cofactor activity; (F) inhibition of C5 convertase activity; (g) inhibition of C3a or C5a production in vitro; (h) inhibition of neutrophil oxidative burst in vitro;
And (i) inhibiting complement-mediated hemolysis in vitro. The composition according to claim 1, having a physiological activity selected from the group consisting of: 3. The soluble complement receptor type 1 molecule is pBSCR1
s, pBM-CR1c, pBSCR1s / pTCSgpt, pT-CR1c1, pT-CR1c2, pT
The composition according to claim 1, which is encoded by a nucleic acid vector selected from the group consisting of -CR1c3, pT-CR1c4 and pT-CR1-c5. 4. The amino acid sequence has the accession number CRL10052 at the ATCC.
2. The composition of claim 1, wherein the composition is encoded by the nucleic acid vector pBSCR1c / pTCSgpt, which is contained in a cell line deposited by E.C. 5. The method according to claim 5, wherein (1) a soluble complement receptor type 1 molecule or a fragment thereof which retains the physiological activity of a full-length complement receptor type 1 molecule, and (2) a thrombolytic agent of human or other animal. A composition comprising an amount effective for use in combination with a method of treating a thrombotic condition. 6. The nucleic acid vector pBSCR1c / pTCSgpt (ATC
C accession number CRL10052), and a soluble complement receptor type 1 molecule having the amino acid sequence encoded by (2) a thrombolytic agent, respectively, for use in combination with a method of treating thrombotic conditions in humans or other animals. A composition comprising an effective amount. 7. A pharmaceutical composition for treating thrombotic conditions or myocardial infarction in humans or other animals, comprising the composition according to claim 1 and a pharmaceutically acceptable carrier or excipient. 8. A soluble complement receptor type 1 molecule (sCR1) or a full-length complement receptor 1 comprising substantially the amino acid sequence shown in FIG. 1 and lacking a transmembrane region.
A fragment containing the bioactivity of the type molecule is contained in a first container, and (2) a thrombolytic agent is contained in a second container, respectively, for use in combination with a method of treating a thrombotic condition in a human or other animal. A pharmaceutical pack for use in the treatment of a thrombotic condition, the amount being an effective amount of 9. A container comprising (1) a soluble complement receptor type 1 molecule or a fragment thereof retaining the biological activity of a full-length complement receptor type 1 molecule in a first container, and (2) a thrombolytic agent in a second container. A pharmaceutical pack for use in treating myocardial infarction, each of which is an amount effective for use in combination with a method for treating a thrombotic condition in a human or other animal. 10. The nucleic acid vector pBSCR1c / pTCSgpt (A
Pharmaceutical pack according to claim 8 or 9, comprising a soluble complement receptor type 1 molecule having an amino acid sequence encoded by TCC Accession No. CRL10052) in a first container and (2) a thrombolytic agent in a second container. (11) the thrombolytic agent is (a) a plasminogen activator or a mutein thereof; (b) an ansoylated plasminogen-streptokinase-activator complex; (c) a single-chain urokinase; (E) streptokinase; (f) a first double-stranded protease bound to one chain of a second double-stranded protease so as to have a catalytic site essential for fibrinolytic activity. A fibrinolytically active hybrid protein comprising one chain, wherein the catalytic site is optionally blocked with a removable blocking group; and (g) a reversibly protected in vivo fibrinolytic enzyme. Wherein the catalytic site essential for fibrinolytic activity in the enzyme has a pseudo-first-order rate constant for hydrolysis of pH 7.4 at 37 ° C. in an isotonic aqueous solution. ^10-6 / sec to 10 ^-3 / seconds range become rate of such selected from the group consisting of the enzymes being blocked by a group which can be removed by hydrolysis, to any one of claims 1-7 A composition as described. (12) the thrombolytic agent is (a) plasminogen activator or its mutein; (b) anisoylated plasminogen-streptokinase-activator complex; (c) single-chain urokinase; (E) streptokinase; (f) a first double-stranded protease bound to one chain of a second double-stranded protease so as to have a catalytic site essential for fibrinolytic activity. A fibrinolytically active hybrid protein comprising one chain, wherein the catalytic site is optionally blocked with a removable blocking group; and (g) a reversibly protected in vivo fibrinolytic enzyme. Wherein the catalytic site essential for fibrinolytic activity in the enzyme has a pseudo-first-order rate constant for hydrolysis of pH 7.4 at 37 ° C. in an isotonic aqueous solution. ^10-6 / sec to 10 ^-3 / seconds range become rate of such selected from the group consisting of the enzymes being blocked by a group which can be removed by hydrolysis, to any one of claims 8 to 10 A pharmaceutical pack as described. 13. Use of the composition according to claim 7, for treating a thrombotic condition. 14. Use of the composition according to claim 7, for the treatment of myocardial infarction.